CA2183232A1 - Remote thermographic sensing of livestock - Google Patents
Remote thermographic sensing of livestockInfo
- Publication number
- CA2183232A1 CA2183232A1 CA002183232A CA2183232A CA2183232A1 CA 2183232 A1 CA2183232 A1 CA 2183232A1 CA 002183232 A CA002183232 A CA 002183232A CA 2183232 A CA2183232 A CA 2183232A CA 2183232 A1 CA2183232 A1 CA 2183232A1
- Authority
- CA
- Canada
- Prior art keywords
- animal
- pixels
- area
- animals
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01K—ANIMAL HUSBANDRY; CARE OF BIRDS, FISHES, INSECTS; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- A01K11/00—Marking of animals
- A01K11/006—Automatic identification systems for animals, e.g. electronic devices, transponders for animals
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01K—ANIMAL HUSBANDRY; CARE OF BIRDS, FISHES, INSECTS; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- A01K29/00—Other apparatus for animal husbandry
- A01K29/005—Monitoring or measuring activity, e.g. detecting heat or mating
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N25/00—Investigating or analyzing materials by the use of thermal means
- G01N25/72—Investigating presence of flaws
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/02—Food
- G01N33/12—Meat; fish
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
- G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
- G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
Abstract
The invention provides a method and apparatus for remote sensing of livestock, using a thermographic image sensing system, in order to determine one or more of the number, weight, location, temperature, carcass pH etc., of animals in a surveillance area. A
thermographic image comprising pixels of the area is sent to a digitizing board in a microcomputer, where the image is converted into a number array. The numbers are then interpreted by software to provide the desired information in a decipherable form.
thermographic image comprising pixels of the area is sent to a digitizing board in a microcomputer, where the image is converted into a number array. The numbers are then interpreted by software to provide the desired information in a decipherable form.
Description
W095/23336 ~1 8 3 2 3 2 PCT/CA95/00093 Remote ThermograPhic Sensing of Livestock The present invention relates to a method and apparatus for the remote sensing of animals using thermographic sensing technology and computer software for analysis and presentation of thermographic data obtained.
Currently in agriculture there are some bottlenecks that impede further advances in production efficiency and animal health and welfare. Growth monitoring is currently done using inventory analysis and physical weighing of animals. These procedures are invasive (ie. disrupt the everyday life and growth of the livestock), labour intensive, and prone to human error.
Advances in growth monitoring would enhance our knowledge of animal production, and would allow producers in the field to identify inefficiency in their operations.
A second major bottleneck is the identification of animals which develop diseases that reduce growth and performance. Currently disease diagnosis depends primarily on human recognition of the signs. This recognition is limited by the experience and ability of the person ex~m;ning the animal and the time-and-labour allotted to screening for disease.
Livestock production would be enhanced if current technology could be refined to do some of the above tasks, eg. with instruments that would remotely sense farm animals and determine their quantity, location, temperature and body mass. Such could be applied to livestock production where accurate animal inventories and daily growth rates are difficult to obtain. Pig farms estimate growth rates over the marketing period (6 months). Poultry farms monitor growth by weighing a random subset of birds on a regular basis. Beef feedlots spend a lot of time and money examining incoming cattle for respiratory disease.
Any advances in automating these processes would be extremely useful. Large swine operations have inventories of many thousands of animals. Any technology that would enhance the monitoring of animals would also be welcomed by the industry.
With the advent of efficient microprocessor technology, remote sensing of objects has been developed for commercial manufacturing processes. Using mechanical and electronic means of determining characteristics such as size, shape, colour, texture, etc. allows for less dependence on human input, and ultimately leads to more efficient handling and processing methods. In some cases, remote sensing and measuring can add a level of precision beyond human capability.
Precise measurements of animals are desired in many livestock operations in order to monitor production.
For example, the weight and temperature of animals are important in assessing production rates, presence of disease, environmental conditions etc. In using traditional techniques, obtaining such measurements involves handling animals individually, which can be manually demanding and time consuming. Ideally, a method of obtaining such measurements remotely, ie. without manually handling each animal individually, and readily upon the demand of a livestock operation worker, would be substantially beneficial.
We have now found a method and apparatus that can be used for remotely sensing animals, in one aspect, a livestock operation and to provide for example, weight and temperature data on demand, or continuously if desired.
These rely in part on thermographic sensing technology.
Thermographic technology has been adapted for many temperature sensing purposes, eg. in medicine for detecting inflammation or other pathological conditions of humans, as disclosed in U.S. Patent No.5,056,525. It has also been used in the diagnosis of pathological conditions in ~1 8-~32 animals, as exemplified in the paper: Hurnik, J.F., De Boer, S., and Webster, A.B., Detection of Health D~orders in Dai~
Cattle Util~ing a ~tem2al In~ed Scanning Techn~ue, Can.J.Anim.Sci. 64:
1071-1073. However, such and other uses to date have been for the analysis of specific animals under specific care or attention, and not performed remotely on animals on a surveillance or continuous basis.
SummarY of the Invention The invention provides an apparatus for remote sensing of information about an animal in an area and making said information available to a person in a decipherable form. The information may be production information about livestock, or health status information of animals in a zoo, for example. The apparatus in a broad aspect comprises a means for remotely obtaining a thermographic image of the area, which image can be used directly or indirectly to provide an absolute or relative temperature of an animal in the area. There is also a means for converting the image into computer readable form, a means for interpreting that form to provide the information, and a means for displaying the information in a person-decipherable form, eg. in a form with meaning to a livestock manager or zoo-keeper.
The means for converting the image into computer readable form preferably includes a means for converting the image into a number array, each number having a - position corresponding to a pixel in the image and a value corresponding to a relative brightness of the pixel. This may be accomplished on a video-digitizer circuit board.
Preferably, the means for remotely obtaining the thermographic image of the area is a thermographic imaging camera. The camera is preferably mountable on a means for moving the camera from over one area to another, so that animals in different areas may be viewed selectively. The woss/23336 ~J 8 3 ~ 3 2 PCT/CA95/00093 means for moving the camera preferably includes a track, extending over two or more different areas. The camera is preferably slideably mounted on the track.
In another preferred embodiment of the apparatus using a thermographic imaging camera, the camera is fixable to view only one area, eg. a temporary holding pen for animals for slaughter. Livestock in such area would be changed regularly, so a fixed camera would be sufficient in such context.
The means for interpreting the computer readable form to provide the information preferably includes processing means for segmenting a group of pixels associated with an animal in the area from pixels associated with background and applying a unique identifier to the animal profile. The processing means also preferably includes one or more of the following: means for segmenting a single group of pixels associated with a plurality of animals and determining an approximation of a number of animals associated with the single group of pixels; means for counting the unique identifier and the approximation of a number of animals associated with the single group of pixels to determine an estimate of a number of animals in the area; means for counting a total number of pixels in an animal profile to determine an approximate weight of an associated animal in the area; means for determining a centre of an animal from locations of all pixels in an associated animal profile and then determining an approximate position of the animal in the area; means for determining an approximate absolute or relative temperature of an animal in the area based on average brightness of pixels in an associated animal profile;
means fcr determining an estimate of carcass 24 hour pH
from the approximate (pixel) temperature variation of an animal before slaughter; means for determining an estimate of visual PSE meat score from the approximate (pixel) temperature of -an animal before slaughter; and means for-woss/23336 ~1 8 3 ~ 3 2 PCTICA95/00093 determining an approximate absolute or relative temperatureof background to animals in the area based on average brightness of pixels in the background.
In another aspect there is provided a method of remotely sensing information about an animal in an area, which comprises obtaining a thermographic image of the area, wherein the image directly or indirectly is able to provide an absolute or relative temperature of an animal in the area, converting the image into computer readable form, interpreting the form to provide the information, and displaying the information in a person-decipherable form.
Brief DescriPtion of the Drawings In the accompanying drawings which illustrate preferred embodiments of the present invention:
FIG. 1 is a perspective view of an apparatus, according to the present invention, in use in monitoring pigs;
FIG. 2 is a graph of actual weight (scale weight) of various pigs vs pixel weight (weight determined from pixel values obtained by thermographic image analysis) in a study of applying the present invention to pigs;
FIG. 3 is a graph of actual skin temperature of various pigs vs pixel temperature ( temperature determined from pixel values obtained by thermographic image analysis) in a study of applying the present invention to pigs;
FIG. 4 is a flow chart of one example of basic thermographic sensing operations that may be employed under the present invention;
FIG. 5 is a graph of actual vs pixel carcass pH in a study of applying the present invention to pigs; and FIG. 6 is a graph of PSE (Pale Soft Exudative meat) score vs pixel temperature in a study of applying the present invention to pigs.
W095/23336 ~1 8 3 2 3 2 PCT/CA95/00093 Detailed DescriPtion of Preferred Embodiments We have developed a method for the remote sensing of animals using a thermographic image sensing system. The method involves sensing the presence of animals, the biomass (number and weight of animals) and temperature of the animals (which is an indicator of the health of the animals). The method is free of human contact or intervention with the animals, and in this sense is ~remote". The information sensed by the sensing system is tabulated and conveyed to a human, eg. on a chart, printout or computer screen.
The thermographic image sensing system uses a thermographic camera which sends a video image to a digitizing board in a microcomputer. Software interprets the following features of the video image: number of animals in the area imaged; mass (estimated weight) of the animals; average skin temperature of each animal;
approximate location of the animals in the imaged area; and the temperature of the surfaces adjoining the location of the animals. The software delivers the above information in a decipherable manner to those gathering information about the animals. Such information is relayed by diskette, wire or infrared transmission from the camera/
digitizing board combination to another computer that is operated by an ~An;r~l handler".
Under the present method, the thermographic screening of livestock before slaughter enables operators of the method to recognize animal health and quality problems while the animal is still alive and/or immediately after slaughter. Thermography detects changes in the body surface temperature as an indication of the metabolic state of the animal. Current non-thermographic meth~ds obtain metabolic status information using invasive,- labour dependant, time consuming techniques. The present method, sometimes referred to as n Remote Thermographic Animal W095/23336 ~ 23 ? PCT/CA95/00093 Sensing" or ~RTAS", allows for automated screening of animals for example, on a farm, or in an abattoir.
Using RTAS, the following conditions are detectable: general body infections or infections to anatomical regions of an animal; injuries to, or bruising of an animal; metabolic disorders such as exhaustion or exertion (eg. Porcine Stress Syndrome (PSS) which can cause pork carcass quality problems such as P.S.E.- Pale Soft Exudative meat condition). Other benefits of using RTAS
include: labour savings to the slaughter and meat processing industries through an automated process for activities currently carried out manually and visually;
improved selection and separation of animals having ideal metabolic status, thereby making the slaughter process more efficient and streamlined; identification of problem animals allows for useful interventions to occur while the animal is alive - eg. an exhausted animal may be rested, or an excited or nervous animal can be allowed to calm down and cool off, or an animal with PSS can be rested which is known to result in improved pork quality; identification of infected animals will allow for removal of those animals for a detailed inspection before slaughter occurs, thereby reducing the risk of disease in consumers, so there is an improvement in food safety and the inspection process is enhanced.
The thermographic sensing system used in the present method requires a thermographic imaging means, eg.
camera, that can alone or with an adjunct, provide absolute or relative temperatures of objects in the field of view of the means. We selected a camera available through the company ISI Group Inc. of 211 Conchas S.E., Albuquerque, New Mexico, U.S.A., namely Model 9100, which operates under pyro-electric vidicon technology and produces a US Standard RS-170 output (30 frames per second). Since-this camera does not provide absolute temperatures, but rather only thermographic images, the images produced required wos~/23336 ~7 8 3 2 3 2 PCT/CA95/00093 calibration to enable relative temperatures and ultimately absolute temperatures to be obtained. Our selected camera required very slight movement to produce a scanned image of the infrared scene.
The microcomputer for selection preferably has the following specifications: 486/33 , or faster, central processing unit; 200 Mb, or higher, hard disk; 10-20 Mb, or higher, ram; ISA bus architecture (due to the selected digitizing board's AT specifications). A digitizing board and software were used to capture and interconnect the imaging camera with the microcomputer and software programs. We selected a Matrox Model IP-8 digitizing board, supplied by Matrox Electronic Systems Ltd.of Dorval, Quebec, Canada, which provides a resolution of 512x512x8 bits, accepts RS-170 input, has a 2 Mb video frame buffer on the card, connects to an ISA bus, and has an extensive library of software development tools and demonstration programs for directly accessing the frame buffers.
Image analysis in the present method is a set of procedures for enhancing the visibility of objects in a scene, separating ~objects of interest", ie. animals, from ~backgroundn, and performing measurements on the objects of interest and, optionally, the background. The scene sensed by the thermographic camera produces a signal which is conducted to a video-digitizer circuit board installed in the microcomputer. The board converts the signal into a square array of numbers inside the computer. The position of a number in the array corresponds to the position of one picture dot, or ~pixel", in the scene. The value of the number indicates the pixel's ~brightness".
The number of pixels in the array determines the resolution of the image. Usually the size of these arrays is 256-rows by 256-columns (more than 65,000 pixels) and may be as large as 1024-rows by 1024-columns (over one million pixels). We used a 512 x 480 size of array in our-Wos~l23336 2 l ~ 3 ~ 3 2 PCT/CA95/00093 g tests. The pixel values typically range from 0 to 255.
The large size of the data array, combined with the intensive computations done on them, preferably requires computer hardware that can perform extremely fast and contain very large amounts of pixel data. Otherwise the time needed to extract information from a scene becomes excessive.
We used a pen with pigs to develop the present method and apparatus (see FIG.1). There is shown in Figure 1 a camera lens 1, start button 6, stop button 7, disc 2 for pen data output, disc 3 for parameter inputs, disc 4 for error trace data output, and report print-out 5.
The thermographic camera was placed above the pen, pointing downwards, so that only the tops of pigs were visible to the camera. Data of interest for collection included the positions, sizes and temperatures of individual pigs.
Computer algorithms for resolving these from the scene were developed and included adjusting for poor contrast between pigs and background, and compensating for overlapping images of pigs, eg. images formed from two or more pigs in substantial contact with each other.
Figure 4 illustrates a flowchart for software designed for use in our remote animal sensing technology.
It will be apparent to any person skilled in the art that other software designs may also be used in accordance with the scope of the present invention, so references to the software in Figure 4 are non-limiting.
In one commercial application of the apparatus of the present invention the camera is designed to move mechanically from animal pen to animal pen. In another application the camera may remain stationary as animals move below and through the field of view. In either case, the first step in the image analysis is to obtain a video frame from the thermographic camera, wherein the frame includes animal images of animals within a predefined field SU~ JTE SHEEr woss/23336 ~ 8 3 2 3 2 pcTlcA9slooos3 of view. In our tests the frame was captured as a 512 x 480 pixel array. The array is then ~cleaned" through an erode and dilate process. The array next undergoes a threshold process to define the animal profiles and separate them from the background ~non-animal" areas. A
row by row method utilizing a fixed formula may be used, as may a method of choosing the threshold by a dip in the histogram that is found at the transition between animal and background pixel values. In some cases further refinements in choosing the threshold value can come from a method of evaluating the image array and choosing the pixel value where the background begins to break up, thereby eliminating the background and not eroding any animal pixels. The final thresholding method may be any one of the above methods either alone or in combination, and is not limited to the described methods, in keeping with the scope of the present invention.
The thresholded image contains defined animal areas which are labelled and analyzed. Objects smaller than a predefined animal area are discarded; objects within an anticipated size are counted; and objects larger than the predetermined size are divided by the predefined size to estimate the number of animals in the large objects.
The latter technique compensates for situations where the animals are so close together that the software perceives them as one animal. The parameter file which contains pen specific information is used to determine the predetermined animal size. The parameter file is initialized whenever the apparatus encounters the pen for the first time. The number of estimated animals is written to the output file, as is the estimation of the temperature of each animal based on the pixel values. The temperature may be in absolute form, ie. C, if the camera is calibrated against an object of known temperature on a regular basis. A
relative determination of the temperature may also be incorporated by noting which animals in each pen have an elevated temperature compared to their pen mates (see Table SUB~ JTE SHEET
Currently in agriculture there are some bottlenecks that impede further advances in production efficiency and animal health and welfare. Growth monitoring is currently done using inventory analysis and physical weighing of animals. These procedures are invasive (ie. disrupt the everyday life and growth of the livestock), labour intensive, and prone to human error.
Advances in growth monitoring would enhance our knowledge of animal production, and would allow producers in the field to identify inefficiency in their operations.
A second major bottleneck is the identification of animals which develop diseases that reduce growth and performance. Currently disease diagnosis depends primarily on human recognition of the signs. This recognition is limited by the experience and ability of the person ex~m;ning the animal and the time-and-labour allotted to screening for disease.
Livestock production would be enhanced if current technology could be refined to do some of the above tasks, eg. with instruments that would remotely sense farm animals and determine their quantity, location, temperature and body mass. Such could be applied to livestock production where accurate animal inventories and daily growth rates are difficult to obtain. Pig farms estimate growth rates over the marketing period (6 months). Poultry farms monitor growth by weighing a random subset of birds on a regular basis. Beef feedlots spend a lot of time and money examining incoming cattle for respiratory disease.
Any advances in automating these processes would be extremely useful. Large swine operations have inventories of many thousands of animals. Any technology that would enhance the monitoring of animals would also be welcomed by the industry.
With the advent of efficient microprocessor technology, remote sensing of objects has been developed for commercial manufacturing processes. Using mechanical and electronic means of determining characteristics such as size, shape, colour, texture, etc. allows for less dependence on human input, and ultimately leads to more efficient handling and processing methods. In some cases, remote sensing and measuring can add a level of precision beyond human capability.
Precise measurements of animals are desired in many livestock operations in order to monitor production.
For example, the weight and temperature of animals are important in assessing production rates, presence of disease, environmental conditions etc. In using traditional techniques, obtaining such measurements involves handling animals individually, which can be manually demanding and time consuming. Ideally, a method of obtaining such measurements remotely, ie. without manually handling each animal individually, and readily upon the demand of a livestock operation worker, would be substantially beneficial.
We have now found a method and apparatus that can be used for remotely sensing animals, in one aspect, a livestock operation and to provide for example, weight and temperature data on demand, or continuously if desired.
These rely in part on thermographic sensing technology.
Thermographic technology has been adapted for many temperature sensing purposes, eg. in medicine for detecting inflammation or other pathological conditions of humans, as disclosed in U.S. Patent No.5,056,525. It has also been used in the diagnosis of pathological conditions in ~1 8-~32 animals, as exemplified in the paper: Hurnik, J.F., De Boer, S., and Webster, A.B., Detection of Health D~orders in Dai~
Cattle Util~ing a ~tem2al In~ed Scanning Techn~ue, Can.J.Anim.Sci. 64:
1071-1073. However, such and other uses to date have been for the analysis of specific animals under specific care or attention, and not performed remotely on animals on a surveillance or continuous basis.
SummarY of the Invention The invention provides an apparatus for remote sensing of information about an animal in an area and making said information available to a person in a decipherable form. The information may be production information about livestock, or health status information of animals in a zoo, for example. The apparatus in a broad aspect comprises a means for remotely obtaining a thermographic image of the area, which image can be used directly or indirectly to provide an absolute or relative temperature of an animal in the area. There is also a means for converting the image into computer readable form, a means for interpreting that form to provide the information, and a means for displaying the information in a person-decipherable form, eg. in a form with meaning to a livestock manager or zoo-keeper.
The means for converting the image into computer readable form preferably includes a means for converting the image into a number array, each number having a - position corresponding to a pixel in the image and a value corresponding to a relative brightness of the pixel. This may be accomplished on a video-digitizer circuit board.
Preferably, the means for remotely obtaining the thermographic image of the area is a thermographic imaging camera. The camera is preferably mountable on a means for moving the camera from over one area to another, so that animals in different areas may be viewed selectively. The woss/23336 ~J 8 3 ~ 3 2 PCT/CA95/00093 means for moving the camera preferably includes a track, extending over two or more different areas. The camera is preferably slideably mounted on the track.
In another preferred embodiment of the apparatus using a thermographic imaging camera, the camera is fixable to view only one area, eg. a temporary holding pen for animals for slaughter. Livestock in such area would be changed regularly, so a fixed camera would be sufficient in such context.
The means for interpreting the computer readable form to provide the information preferably includes processing means for segmenting a group of pixels associated with an animal in the area from pixels associated with background and applying a unique identifier to the animal profile. The processing means also preferably includes one or more of the following: means for segmenting a single group of pixels associated with a plurality of animals and determining an approximation of a number of animals associated with the single group of pixels; means for counting the unique identifier and the approximation of a number of animals associated with the single group of pixels to determine an estimate of a number of animals in the area; means for counting a total number of pixels in an animal profile to determine an approximate weight of an associated animal in the area; means for determining a centre of an animal from locations of all pixels in an associated animal profile and then determining an approximate position of the animal in the area; means for determining an approximate absolute or relative temperature of an animal in the area based on average brightness of pixels in an associated animal profile;
means fcr determining an estimate of carcass 24 hour pH
from the approximate (pixel) temperature variation of an animal before slaughter; means for determining an estimate of visual PSE meat score from the approximate (pixel) temperature of -an animal before slaughter; and means for-woss/23336 ~1 8 3 ~ 3 2 PCTICA95/00093 determining an approximate absolute or relative temperatureof background to animals in the area based on average brightness of pixels in the background.
In another aspect there is provided a method of remotely sensing information about an animal in an area, which comprises obtaining a thermographic image of the area, wherein the image directly or indirectly is able to provide an absolute or relative temperature of an animal in the area, converting the image into computer readable form, interpreting the form to provide the information, and displaying the information in a person-decipherable form.
Brief DescriPtion of the Drawings In the accompanying drawings which illustrate preferred embodiments of the present invention:
FIG. 1 is a perspective view of an apparatus, according to the present invention, in use in monitoring pigs;
FIG. 2 is a graph of actual weight (scale weight) of various pigs vs pixel weight (weight determined from pixel values obtained by thermographic image analysis) in a study of applying the present invention to pigs;
FIG. 3 is a graph of actual skin temperature of various pigs vs pixel temperature ( temperature determined from pixel values obtained by thermographic image analysis) in a study of applying the present invention to pigs;
FIG. 4 is a flow chart of one example of basic thermographic sensing operations that may be employed under the present invention;
FIG. 5 is a graph of actual vs pixel carcass pH in a study of applying the present invention to pigs; and FIG. 6 is a graph of PSE (Pale Soft Exudative meat) score vs pixel temperature in a study of applying the present invention to pigs.
W095/23336 ~1 8 3 2 3 2 PCT/CA95/00093 Detailed DescriPtion of Preferred Embodiments We have developed a method for the remote sensing of animals using a thermographic image sensing system. The method involves sensing the presence of animals, the biomass (number and weight of animals) and temperature of the animals (which is an indicator of the health of the animals). The method is free of human contact or intervention with the animals, and in this sense is ~remote". The information sensed by the sensing system is tabulated and conveyed to a human, eg. on a chart, printout or computer screen.
The thermographic image sensing system uses a thermographic camera which sends a video image to a digitizing board in a microcomputer. Software interprets the following features of the video image: number of animals in the area imaged; mass (estimated weight) of the animals; average skin temperature of each animal;
approximate location of the animals in the imaged area; and the temperature of the surfaces adjoining the location of the animals. The software delivers the above information in a decipherable manner to those gathering information about the animals. Such information is relayed by diskette, wire or infrared transmission from the camera/
digitizing board combination to another computer that is operated by an ~An;r~l handler".
Under the present method, the thermographic screening of livestock before slaughter enables operators of the method to recognize animal health and quality problems while the animal is still alive and/or immediately after slaughter. Thermography detects changes in the body surface temperature as an indication of the metabolic state of the animal. Current non-thermographic meth~ds obtain metabolic status information using invasive,- labour dependant, time consuming techniques. The present method, sometimes referred to as n Remote Thermographic Animal W095/23336 ~ 23 ? PCT/CA95/00093 Sensing" or ~RTAS", allows for automated screening of animals for example, on a farm, or in an abattoir.
Using RTAS, the following conditions are detectable: general body infections or infections to anatomical regions of an animal; injuries to, or bruising of an animal; metabolic disorders such as exhaustion or exertion (eg. Porcine Stress Syndrome (PSS) which can cause pork carcass quality problems such as P.S.E.- Pale Soft Exudative meat condition). Other benefits of using RTAS
include: labour savings to the slaughter and meat processing industries through an automated process for activities currently carried out manually and visually;
improved selection and separation of animals having ideal metabolic status, thereby making the slaughter process more efficient and streamlined; identification of problem animals allows for useful interventions to occur while the animal is alive - eg. an exhausted animal may be rested, or an excited or nervous animal can be allowed to calm down and cool off, or an animal with PSS can be rested which is known to result in improved pork quality; identification of infected animals will allow for removal of those animals for a detailed inspection before slaughter occurs, thereby reducing the risk of disease in consumers, so there is an improvement in food safety and the inspection process is enhanced.
The thermographic sensing system used in the present method requires a thermographic imaging means, eg.
camera, that can alone or with an adjunct, provide absolute or relative temperatures of objects in the field of view of the means. We selected a camera available through the company ISI Group Inc. of 211 Conchas S.E., Albuquerque, New Mexico, U.S.A., namely Model 9100, which operates under pyro-electric vidicon technology and produces a US Standard RS-170 output (30 frames per second). Since-this camera does not provide absolute temperatures, but rather only thermographic images, the images produced required wos~/23336 ~7 8 3 2 3 2 PCT/CA95/00093 calibration to enable relative temperatures and ultimately absolute temperatures to be obtained. Our selected camera required very slight movement to produce a scanned image of the infrared scene.
The microcomputer for selection preferably has the following specifications: 486/33 , or faster, central processing unit; 200 Mb, or higher, hard disk; 10-20 Mb, or higher, ram; ISA bus architecture (due to the selected digitizing board's AT specifications). A digitizing board and software were used to capture and interconnect the imaging camera with the microcomputer and software programs. We selected a Matrox Model IP-8 digitizing board, supplied by Matrox Electronic Systems Ltd.of Dorval, Quebec, Canada, which provides a resolution of 512x512x8 bits, accepts RS-170 input, has a 2 Mb video frame buffer on the card, connects to an ISA bus, and has an extensive library of software development tools and demonstration programs for directly accessing the frame buffers.
Image analysis in the present method is a set of procedures for enhancing the visibility of objects in a scene, separating ~objects of interest", ie. animals, from ~backgroundn, and performing measurements on the objects of interest and, optionally, the background. The scene sensed by the thermographic camera produces a signal which is conducted to a video-digitizer circuit board installed in the microcomputer. The board converts the signal into a square array of numbers inside the computer. The position of a number in the array corresponds to the position of one picture dot, or ~pixel", in the scene. The value of the number indicates the pixel's ~brightness".
The number of pixels in the array determines the resolution of the image. Usually the size of these arrays is 256-rows by 256-columns (more than 65,000 pixels) and may be as large as 1024-rows by 1024-columns (over one million pixels). We used a 512 x 480 size of array in our-Wos~l23336 2 l ~ 3 ~ 3 2 PCT/CA95/00093 g tests. The pixel values typically range from 0 to 255.
The large size of the data array, combined with the intensive computations done on them, preferably requires computer hardware that can perform extremely fast and contain very large amounts of pixel data. Otherwise the time needed to extract information from a scene becomes excessive.
We used a pen with pigs to develop the present method and apparatus (see FIG.1). There is shown in Figure 1 a camera lens 1, start button 6, stop button 7, disc 2 for pen data output, disc 3 for parameter inputs, disc 4 for error trace data output, and report print-out 5.
The thermographic camera was placed above the pen, pointing downwards, so that only the tops of pigs were visible to the camera. Data of interest for collection included the positions, sizes and temperatures of individual pigs.
Computer algorithms for resolving these from the scene were developed and included adjusting for poor contrast between pigs and background, and compensating for overlapping images of pigs, eg. images formed from two or more pigs in substantial contact with each other.
Figure 4 illustrates a flowchart for software designed for use in our remote animal sensing technology.
It will be apparent to any person skilled in the art that other software designs may also be used in accordance with the scope of the present invention, so references to the software in Figure 4 are non-limiting.
In one commercial application of the apparatus of the present invention the camera is designed to move mechanically from animal pen to animal pen. In another application the camera may remain stationary as animals move below and through the field of view. In either case, the first step in the image analysis is to obtain a video frame from the thermographic camera, wherein the frame includes animal images of animals within a predefined field SU~ JTE SHEEr woss/23336 ~ 8 3 2 3 2 pcTlcA9slooos3 of view. In our tests the frame was captured as a 512 x 480 pixel array. The array is then ~cleaned" through an erode and dilate process. The array next undergoes a threshold process to define the animal profiles and separate them from the background ~non-animal" areas. A
row by row method utilizing a fixed formula may be used, as may a method of choosing the threshold by a dip in the histogram that is found at the transition between animal and background pixel values. In some cases further refinements in choosing the threshold value can come from a method of evaluating the image array and choosing the pixel value where the background begins to break up, thereby eliminating the background and not eroding any animal pixels. The final thresholding method may be any one of the above methods either alone or in combination, and is not limited to the described methods, in keeping with the scope of the present invention.
The thresholded image contains defined animal areas which are labelled and analyzed. Objects smaller than a predefined animal area are discarded; objects within an anticipated size are counted; and objects larger than the predetermined size are divided by the predefined size to estimate the number of animals in the large objects.
The latter technique compensates for situations where the animals are so close together that the software perceives them as one animal. The parameter file which contains pen specific information is used to determine the predetermined animal size. The parameter file is initialized whenever the apparatus encounters the pen for the first time. The number of estimated animals is written to the output file, as is the estimation of the temperature of each animal based on the pixel values. The temperature may be in absolute form, ie. C, if the camera is calibrated against an object of known temperature on a regular basis. A
relative determination of the temperature may also be incorporated by noting which animals in each pen have an elevated temperature compared to their pen mates (see Table SUB~ JTE SHEET
2 1 83~32 wossl23336 PCT/CA95/00093 VII).
- An average of the non-animal pixel values represents the background pen temperatures. These are written to the output file either in absolute or relative form.
The X and Y coordinates of the centre of each object will give the location of each animal and this information is then written to the output file.
The output file information is processed by a database management program to deliver information to the livestock operator in a useful manner. Inventory information can provide an estimate of the total animal inventory. Warning indicators can notify the livestock operator if significant changes in inventory are found.
Animal weight data can be used to provide growth rate estimates for each pen of animals. Warning indicators can notify the operator if the animals are not growing at the optimal rates or if the animals are about to reach market weights. The disease incidence rate can be reported for each pen by noting the number of animals with elevated temperatures (fevers), again warning indicators can alert the livestock operator of disease outbreaks and which pens are affected. The location and background information can be plotted in graphic form to allow the producer to view the locations of the animals and remotely view the temperature status of the pens. Warning indicators can draw the operator's attention to pens which are abnormally cold or hot.
our work has also established a link between thermographic remote sensing of pigs prior to slaughter and subsequent carcass meat quality. Carcass 24 hour pH and visual PSE meat score can be predicted from temperature data obtained under the present invention. With respect to pH, it is the case that the pH of pork 24 hours after SUB~ JTE SHEEl-~ 1 83232 slaughter is a criterion of meat quality. We have found that animals with a large variation in pixel values are associated with an increased muscle pH (see Figure 5).
This relationship is strengthened when looking at pigs within on farm source. The significance of this finding is that commercial abattoirs may be able to sort loads of pigs on arrival into those that will give the highest quality pork.
The PSE condition of pork is a major quality flaw resulting in significant marketing losses. Thermographic sensing of pigs prior to slaughter has revealed a relationship between PSE pork and the thermographic pixel values (see Figure 6). Pigs with higher pixel values tend to have an elevated PSE score, particularly when the score is greater than 2. Detecting PSE problems prior to slaughter can allow for quality selection when sorting live pigs and diverting PSE pigs to a rest facility where the pigs can rest and reduce the incidence of PSE pork.
In summary of some of the preferred aspects of one embodiment of the present invention, separating ~pig"
pixels into individual pigs involves scanning the image row-by-row starting from the top; when a pig profile is encountered, it is labelled with a unique identifier.
Scanning continues until every pig profile has been filled with a unique identifier. Compensation for overlapping pig profiles is made. Once labelled, the unique identifiers and the pixels for each unique identifier may be used to determine one or more of the following:
Numbers: The unique identifiers yield an approximation of the number of pigs in the pen;
Weights: A count of the number of pixels in each unique identifier yields a measurement of the mass of the pig;
Positions: Calculating the center for each identifier yields where the pigs are positioned in the SUB~ ~ TE SHEET
pen;
Temperatures: Averaging the brightness of the pixels associated with each identifier yields a measure of the pig's body temperature. Pixel - 5 temperature can be used to predict actual carcass 24 hour pH and visual PSE meat score;
and Background: Averaging the brightness of the pixels not associated with any identifier, ie. of the ~non-pig" pixels, approximates the background (pen) temperature.
There may be other useful kinds of information that can be obtained from the thermographic image data, as may occur to someone skilled in the art from consideration of the present description, so the above list is not intended to be limiting.
The following Tables I-III illustrate a Parameter File layout and a sample Parameter File which have been developed for use in preferred software for carrying out the method and apparatus of the subject invention.
,/
/
SUBS 111 ~JTE SHEE~
~ ~ 83~3~
wos~/23336 PCT/CA95/00093 TABLE I
Parameter File LaYout DEFAULT
KEYWORD VALUE ACCEPTED VALUES
Barn-Name= blank one-line phrase to identify/describe the barn;
used as a title on any printouts Barn-ID= 'BARN' 4 character mnemonic uniquely identifying this barn location Barn-Temp-MAX= 15 Range for 'background' barn Barn-Temp-MIN= 10 temperatures Species= Porcine Porcine, Bovine, etc Shipping-Weight= 102 target shipping weight (see Weight-Units) Normal-Body- 39 normal average body Temp= temperature (see Temp-Units) Temp-Units= C units of measure for all TEMPERATURE measurements;
C=celsius; F=fahrenheit;
K=kelvin Distance-Units= M units of measure for all DISTANCE measurements;
R=screen rastor units;
MM=millimeters;
CM=centimeters; M=meters;
IN=inches; FT=feet;
YD=yards Weight-Units= KG units of measure for all WEIGHT measurements;
GM=grams; KG=kilograms;
OZ=ounces; LB=pounds SUB~ JTE SHEET
DEFAULT
KEYWORD VALUE ACCEPTED VALUES
Lens= 18mm l ens installed in camera Output-Filename= Cymmddhh 8 character file name AND 3 .mm1 character file extension to identify the barn DATA
file;
Printing= S D(etails only) S(ummaries only) or B(oth): indicates whether a printout is required; necessitates a printing device being correctly attached Example "C2123123.59" where y= last digit of year eg.2 is 92; mm= month eg.l2; dd= day eg.31; hh=
hour eg.23; mm= minute eg.59 SUBSTITUTE SHEE~
~ 1 83232 TABLE II
Parameter File LaYout DEFAULT
KEYWORD VALUE ACCEPTED VALUES
The following parameters are repeated for each pen.
Pen-Width= 20 width of the pen (see Distance-Units) Pen-Length= 20 length of the pen (see Distance-Units) Pen-Camera- 10 height of camera above the Height= pen (see Distance-Units) Pen-Avg-Weight average weight of an animal to be housed in this pen;
used to 'separate' overlapping images when performing the inventory counting Pen-MAX-Temp= Range for average animal Pen-MIN-Temp= body temperature in this pen Pen-MAX-Weight= Range for average animal Pen-MIN-Weight= body weight in this pen SUB~ lTE SHEET
~ 1 83232 DEFAULT
KEYWORD VALUE ACCEPTED VALUES
Pen-I D= A (8) 8-characters of alpha-numerics labelling each pen in the order they are 'seen'; upon reading this parameter, the program pauses until 'signalled'2 to proceed;
2 'Signalled' during research and development means that the operator will 'Press Any Key to Continue'; however, for the final product, some form of "trip switch" will likely indicate that the camera is in the correct position above - the pen and that the program should proceed wi-th it's computations.
SUt~;3 111 ~JTE SHEET
W095/23336 ~ 1 8 3 2 3 2 PCT/CA95/00093 TABLE III
Sample Parameter File Barn-Name=Swineman, Joe Barn-ID=JOES
Species=PORCINE
Shipping-Weight=85 Normal-Body-Temp=39 Temp-Units=C
Distance-Units=FT
Weight-Units=KG
Lens=18 Output-Filename=SWINEMAN.JOE
Printing=Y
Pen-Width=17.5 Pen-Length=15.3 Pen-Camera-Height=12.3 Pen-ID=002 Pen-Width=14.4 Pen-Length=18.75 Pen-Camera-Height=12.3 Pen-ID=001 Pen-Width=lS.25 Pen-Length=15.75 Pen-Camera-Height=14.75 Pen-ID=004 Pen-Width=18.8 Pen-Length=20.25 Pen-Camera-Height=14.75 Pen-ID=003 SUB;~ JTE SHEEr WO9Sl23336 PCT/CA95/00093 Table IV below illustrates an output data file in ASCII text, one row per animal per pen, columnar data / /
/
SUB~ JTE SHEEl WO 95/23336 2 i 8 3 ~ 3 2 PCT/CA95/00093 ~ Z L
.,1 ~ ~ ~
O~ ` L
o a~ ~ X
-1 C a~ ` G~ ~ C.) ~1 0 ~ O O O O ~ o O ~ ~ a o m ~: ~n ~n E~ ~ JJ n ~ n a ~
~ H :~ ZC~
_1 ~ z ~ a~ ~ 3 ~ ~ ~ a~
-1 ~ a~ n a~ ~a~ ~ o a~ s~ a _ m ~ 3 ~ E~ X ~ U o + u N
U~
V
SUB~ 111 lJTE SHEEr e ~_ O ~ ~7 N ~ 1-~ ;
N -- ~ ~ ~--~ ,, ~ .. .. . C
N 1/~ ~ N r-~ U
O OO Lô `0 ~ V~
OOOOO
- - - - C C
~ U
N N ~ ~ _ X U~
U~ O O g O O o O O O O O ~ ' C
U U U U U ~
~ N ~ -- e ~
o o o o o ~ y '' '' ~ ........ .. ..... : ~
`O N ro `S `O ~E
r-l O o o o . .E
O O O O O ...... ~ ' '' '' e N r~7 ~ Ul ' g g g $ g .- - E
N L~ 0 0 c~
N o o7 o ~ r7 ~o al _ _ _ _ _ O
n N N N N N
O~ O` O~ O~ O~ .......... - ............. ~
t~ N N N N Z
OCD ~ CSl c~ m c _ o e e e e e cJ c~
Z 2 Z z z G --~ o o o o o a rt-- u u u ~ ~1 X u s ~ ~ ~e ~ .... ,, ,, ., ,, ,, .. , ,, ...... :
SUBS 111 ~JTE SHEET
2 ~ 8~232 Table V below illustrates Field Definitions developed for use in preferred software for use in carrying out the method and apparatus of the subject invention:
SUB~ TE SHEEr ~ 1 83~32 wossl23336 PCT/CA95/00093 TABLE V
Field Definitions - 1. Barn ID ....... Barn Identification: 4-character mnemonic to identify the barn in which the following pens are located 2. Pen Num ....... Pen Number: identification label assigned to uniquely identify each pen; obtained from Parameter File
- An average of the non-animal pixel values represents the background pen temperatures. These are written to the output file either in absolute or relative form.
The X and Y coordinates of the centre of each object will give the location of each animal and this information is then written to the output file.
The output file information is processed by a database management program to deliver information to the livestock operator in a useful manner. Inventory information can provide an estimate of the total animal inventory. Warning indicators can notify the livestock operator if significant changes in inventory are found.
Animal weight data can be used to provide growth rate estimates for each pen of animals. Warning indicators can notify the operator if the animals are not growing at the optimal rates or if the animals are about to reach market weights. The disease incidence rate can be reported for each pen by noting the number of animals with elevated temperatures (fevers), again warning indicators can alert the livestock operator of disease outbreaks and which pens are affected. The location and background information can be plotted in graphic form to allow the producer to view the locations of the animals and remotely view the temperature status of the pens. Warning indicators can draw the operator's attention to pens which are abnormally cold or hot.
our work has also established a link between thermographic remote sensing of pigs prior to slaughter and subsequent carcass meat quality. Carcass 24 hour pH and visual PSE meat score can be predicted from temperature data obtained under the present invention. With respect to pH, it is the case that the pH of pork 24 hours after SUB~ JTE SHEEl-~ 1 83232 slaughter is a criterion of meat quality. We have found that animals with a large variation in pixel values are associated with an increased muscle pH (see Figure 5).
This relationship is strengthened when looking at pigs within on farm source. The significance of this finding is that commercial abattoirs may be able to sort loads of pigs on arrival into those that will give the highest quality pork.
The PSE condition of pork is a major quality flaw resulting in significant marketing losses. Thermographic sensing of pigs prior to slaughter has revealed a relationship between PSE pork and the thermographic pixel values (see Figure 6). Pigs with higher pixel values tend to have an elevated PSE score, particularly when the score is greater than 2. Detecting PSE problems prior to slaughter can allow for quality selection when sorting live pigs and diverting PSE pigs to a rest facility where the pigs can rest and reduce the incidence of PSE pork.
In summary of some of the preferred aspects of one embodiment of the present invention, separating ~pig"
pixels into individual pigs involves scanning the image row-by-row starting from the top; when a pig profile is encountered, it is labelled with a unique identifier.
Scanning continues until every pig profile has been filled with a unique identifier. Compensation for overlapping pig profiles is made. Once labelled, the unique identifiers and the pixels for each unique identifier may be used to determine one or more of the following:
Numbers: The unique identifiers yield an approximation of the number of pigs in the pen;
Weights: A count of the number of pixels in each unique identifier yields a measurement of the mass of the pig;
Positions: Calculating the center for each identifier yields where the pigs are positioned in the SUB~ ~ TE SHEET
pen;
Temperatures: Averaging the brightness of the pixels associated with each identifier yields a measure of the pig's body temperature. Pixel - 5 temperature can be used to predict actual carcass 24 hour pH and visual PSE meat score;
and Background: Averaging the brightness of the pixels not associated with any identifier, ie. of the ~non-pig" pixels, approximates the background (pen) temperature.
There may be other useful kinds of information that can be obtained from the thermographic image data, as may occur to someone skilled in the art from consideration of the present description, so the above list is not intended to be limiting.
The following Tables I-III illustrate a Parameter File layout and a sample Parameter File which have been developed for use in preferred software for carrying out the method and apparatus of the subject invention.
,/
/
SUBS 111 ~JTE SHEE~
~ ~ 83~3~
wos~/23336 PCT/CA95/00093 TABLE I
Parameter File LaYout DEFAULT
KEYWORD VALUE ACCEPTED VALUES
Barn-Name= blank one-line phrase to identify/describe the barn;
used as a title on any printouts Barn-ID= 'BARN' 4 character mnemonic uniquely identifying this barn location Barn-Temp-MAX= 15 Range for 'background' barn Barn-Temp-MIN= 10 temperatures Species= Porcine Porcine, Bovine, etc Shipping-Weight= 102 target shipping weight (see Weight-Units) Normal-Body- 39 normal average body Temp= temperature (see Temp-Units) Temp-Units= C units of measure for all TEMPERATURE measurements;
C=celsius; F=fahrenheit;
K=kelvin Distance-Units= M units of measure for all DISTANCE measurements;
R=screen rastor units;
MM=millimeters;
CM=centimeters; M=meters;
IN=inches; FT=feet;
YD=yards Weight-Units= KG units of measure for all WEIGHT measurements;
GM=grams; KG=kilograms;
OZ=ounces; LB=pounds SUB~ JTE SHEET
DEFAULT
KEYWORD VALUE ACCEPTED VALUES
Lens= 18mm l ens installed in camera Output-Filename= Cymmddhh 8 character file name AND 3 .mm1 character file extension to identify the barn DATA
file;
Printing= S D(etails only) S(ummaries only) or B(oth): indicates whether a printout is required; necessitates a printing device being correctly attached Example "C2123123.59" where y= last digit of year eg.2 is 92; mm= month eg.l2; dd= day eg.31; hh=
hour eg.23; mm= minute eg.59 SUBSTITUTE SHEE~
~ 1 83232 TABLE II
Parameter File LaYout DEFAULT
KEYWORD VALUE ACCEPTED VALUES
The following parameters are repeated for each pen.
Pen-Width= 20 width of the pen (see Distance-Units) Pen-Length= 20 length of the pen (see Distance-Units) Pen-Camera- 10 height of camera above the Height= pen (see Distance-Units) Pen-Avg-Weight average weight of an animal to be housed in this pen;
used to 'separate' overlapping images when performing the inventory counting Pen-MAX-Temp= Range for average animal Pen-MIN-Temp= body temperature in this pen Pen-MAX-Weight= Range for average animal Pen-MIN-Weight= body weight in this pen SUB~ lTE SHEET
~ 1 83232 DEFAULT
KEYWORD VALUE ACCEPTED VALUES
Pen-I D= A (8) 8-characters of alpha-numerics labelling each pen in the order they are 'seen'; upon reading this parameter, the program pauses until 'signalled'2 to proceed;
2 'Signalled' during research and development means that the operator will 'Press Any Key to Continue'; however, for the final product, some form of "trip switch" will likely indicate that the camera is in the correct position above - the pen and that the program should proceed wi-th it's computations.
SUt~;3 111 ~JTE SHEET
W095/23336 ~ 1 8 3 2 3 2 PCT/CA95/00093 TABLE III
Sample Parameter File Barn-Name=Swineman, Joe Barn-ID=JOES
Species=PORCINE
Shipping-Weight=85 Normal-Body-Temp=39 Temp-Units=C
Distance-Units=FT
Weight-Units=KG
Lens=18 Output-Filename=SWINEMAN.JOE
Printing=Y
Pen-Width=17.5 Pen-Length=15.3 Pen-Camera-Height=12.3 Pen-ID=002 Pen-Width=14.4 Pen-Length=18.75 Pen-Camera-Height=12.3 Pen-ID=001 Pen-Width=lS.25 Pen-Length=15.75 Pen-Camera-Height=14.75 Pen-ID=004 Pen-Width=18.8 Pen-Length=20.25 Pen-Camera-Height=14.75 Pen-ID=003 SUB;~ JTE SHEEr WO9Sl23336 PCT/CA95/00093 Table IV below illustrates an output data file in ASCII text, one row per animal per pen, columnar data / /
/
SUB~ JTE SHEEl WO 95/23336 2 i 8 3 ~ 3 2 PCT/CA95/00093 ~ Z L
.,1 ~ ~ ~
O~ ` L
o a~ ~ X
-1 C a~ ` G~ ~ C.) ~1 0 ~ O O O O ~ o O ~ ~ a o m ~: ~n ~n E~ ~ JJ n ~ n a ~
~ H :~ ZC~
_1 ~ z ~ a~ ~ 3 ~ ~ ~ a~
-1 ~ a~ n a~ ~a~ ~ o a~ s~ a _ m ~ 3 ~ E~ X ~ U o + u N
U~
V
SUB~ 111 lJTE SHEEr e ~_ O ~ ~7 N ~ 1-~ ;
N -- ~ ~ ~--~ ,, ~ .. .. . C
N 1/~ ~ N r-~ U
O OO Lô `0 ~ V~
OOOOO
- - - - C C
~ U
N N ~ ~ _ X U~
U~ O O g O O o O O O O O ~ ' C
U U U U U ~
~ N ~ -- e ~
o o o o o ~ y '' '' ~ ........ .. ..... : ~
`O N ro `S `O ~E
r-l O o o o . .E
O O O O O ...... ~ ' '' '' e N r~7 ~ Ul ' g g g $ g .- - E
N L~ 0 0 c~
N o o7 o ~ r7 ~o al _ _ _ _ _ O
n N N N N N
O~ O` O~ O~ O~ .......... - ............. ~
t~ N N N N Z
OCD ~ CSl c~ m c _ o e e e e e cJ c~
Z 2 Z z z G --~ o o o o o a rt-- u u u ~ ~1 X u s ~ ~ ~e ~ .... ,, ,, ., ,, ,, .. , ,, ...... :
SUBS 111 ~JTE SHEET
2 ~ 8~232 Table V below illustrates Field Definitions developed for use in preferred software for use in carrying out the method and apparatus of the subject invention:
SUB~ TE SHEEr ~ 1 83~32 wossl23336 PCT/CA95/00093 TABLE V
Field Definitions - 1. Barn ID ....... Barn Identification: 4-character mnemonic to identify the barn in which the following pens are located 2. Pen Num ....... Pen Number: identification label assigned to uniquely identify each pen; obtained from Parameter File
3. Date ..... Date: format YYMMDD; eg. 921203 is Dec. 3, 1992
4. Time ..... Time of Day: 24-hour clock;
format HHMM; eg. 2143 is 21 hours and 43 minutes (ie.
9:43pm)
format HHMM; eg. 2143 is 21 hours and 43 minutes (ie.
9:43pm)
5. Anim Num . . . Animal Number: sequential number counting each labelled image; this is not an animal identification number but rather a sequential count of the numbers of animals in the pen
6. Est Weight . . Estimated Weight:
7. Weight Units . Units for Estimated Weight:
GM= grams, KG= kilograms, OZ= ounces, LB= pounds
GM= grams, KG= kilograms, OZ= ounces, LB= pounds
8. Avg Temp . . . Average Temperature:
9. Temp Units . . Units for Average Temperature:
C= Celsius, F= Fahrenheit, K= Kelvin
C= Celsius, F= Fahrenheit, K= Kelvin
10. X center . . . X coordinate for Center of Gravity
11. Y center . . . Y coordinate for Center- of Gravity
12. Center Units . Units for Center of Gravity: same SUBS 111 I.JTE SHEET
2 ~ 8~32 units for both X and Y
coordinates; R= rastor, M M = m i l l i m e t e r s , CM= centimeters, M= meters, IN= inches, FT= feet3. Orient Ang . . Angle of Orientation: units are "degrees" of rotation from zero; zero is pointing to right of image screen '~ I 83232 woss/23336 rcTlcAs5looos3 Table VI below illustrates a typical Pen Details Report that may be printed at the end of each pen, ie. as an available report layout under operation of the preferred software. This ~ticker tape" type of report would provide - 5 immediate feedback and information to the herdsman or farm manager, eg. while the unit is proceeding through a barn.
The latter example anticipates that the camera would be on an overhead guide or track and could be selectively moved from one area for surveillance to another.
SUB~ 111 UTE SHEEI' ~ 26 --a~ -a~
,~
a ~
~ Ei -- h ~n '~ h .~ O
v ~ -- ~
r ~ 2 Y O` ~ o ., s S ~ ~ ~ x ~ _ U '1) ~ E
a~ ~ " u~ "
a 3 o ~ ,~
O ~ ~ E
'{5 0 ~ x ~' U
a~
h h ~ r~
h ~ 5~ e N
v ~
ta u _ N U N r~l ~ ~ S ' ~ V ~
P~ a' ~ 8 a~ x N ~ ~ _ a~ a~
h ~
~ ~ EE o~ ~ ~ _ h ~ _ ~- 2 ~~ r~
0 ~ ~J~ 2 e ~c~ ~ N ~ ' a) a) ~ 2 .. .. 2 ~_ 3 N N ~1 N
h ~ l~ - c _ C ~ -- N N N
0 .,1 ~ I J 1~ G -- O O O O
V U
U
a) ~
3 ~s a, ,~ J
U P~ H
a) ~ a ~ -p~ ~ h ~
SUB~ 111 ~JTE SHEET
~ I ~ 3 ~ 3 2 C~ ~ V V~ ~ ., V C~
0 ~ ~ 0 C~ o ~ 0 ~
._ . --. _ ._ , , ~ ._ , , xv~ -- Q x v~ _ E x ~n _ E
o Q ~_ ~ _ Q ~_ ~ Q ~ o, Q 0 E~ v~ E 0 E
~Y ~ 8 ~E
X X X
U U~ U
E
- _ G u~ u~ IZI u~
C ~ C~J -- ~ ~ C
e ~ _ eN _ e X N t~i r~ -- X~ i ~ -- X ~`J
Q Q Q
E ~ ~ ~ E O` ~ ~ E O`
~ - 0 C>~ 0 ~ f~ ~C~ 0 0 ~ . ~-- 0 c~ ~ v ~ v ~ c ~ v m ~ v z c~ ~n ~ ~ -- 2 Cl u~ r7 ~ -- z Cl v ~ 0 ~-- 0 C:1 ~ 1. ~ O` 0 0 0 u ~J o o -- -- u ~J O O -- --Cl N O
-- O o O O ~ -- OO O O G _ o - SUBS~ 1TE SHEE~
W095/23336 2 1 8 3 ~ 3 2 PCT/CA95/00093 Table VII below illustrates a Summary Report that may be printed at the end of the entire circuit of pens.
The livestock manager can therefore be provided with a condensed, pen-based, inventory showing total numbers of pigs: in each pen; their temperatures; and, their weights for comparison to desired shipping weights for exampl 7 SUB~ 111 UTE SHEET
WO95t23336 ~1~ 3 2 3 2 pcTlcAsslooo93 TABLE VII
Summary Report S U M M A R Y
_______ Barn-Name: Swineman, Joe Species: PORCINE
Lens: 18mm** Shipping-Weight: 105KG
File: SWINEMAN.JOE* Normal-Body-Temp: 39C
March 12, 1993 # with # at Total#
Time PenHiTemp ShipWt in Pen _______ ________ ______ ______ 9:34pm PEN2AB-2 3 4 25 9:39pm PEN2AB-1 2 5 15 9:44pm PEN2AB-4 0 2 13 9:49pm PEN2AB-3 1 0 19 SU~;~ JTE SHEEl' wos~/23336 2 1 8 3 2 3 2 PCT/CA95/00093 The preferable external interface for the processed data under the present method and apparatus is the data set of pen-based observations. This interface could also be achieved by a removable floppy diskette for transporting the data set(s) to a microcomputer for use in the database management program. It is also contemplated that the data set could be transmitted in other ways such as wireless modem or hardwired asynchronous communications line.
FIG. 2 indicates that estimating the weight of pigs under the present method has been achieved in our experiments. Plotting actual weight (scale weight - Y
axis) vs pixel value (X axis) shows a strong, useful relationship exists. Although less data is available for correlating average pixel brightness to actual pig temperature, FIG. 3 does establish that a useful relationship exists between such data and that broad estimates of an animal's temperature can be measured remotely. Our tests also indicate that thermographic remote sensing appears to be able to predict pig carcass pH, as shown in Table VIII below and in Figure 5.
/
SU8:~ 1 1 1 UTE SHER
WO 95/23336 2 1 3 ~ ~ 3 2 PCT/CA95/00093 _I
~1 o g :, n a~
~3 ~ ,~
o o~
o ~ ~ ~ o 11 ~ o o o o t H ~1 0 CO
U~ O O O O
U.
~ I U
n ~V ~
uq ~ t~ 1 11 r o O
E~ ~ ~ ~ ~ u ~ ~ ~q 3 u ~ ~ ~
~ V
H L~ V ~V
H ~q _, H~q ~ ~ w ~ ,r ~ ~I Ln ~ O_I ~ Ln Lr~ ~
~r Ln ~1 ~ ~ ~ ~
. Ln o o 1~ ~ ~ Ln o ~:~ o o o o ~V a~ ~ o o U~ O O O O r~ ~ E O O
h ~
o 11 ~ h o o a o ~ ~ Ln 1`
uq E~ 5: C~ N ~ ~ ~D
~ Z -- ~V ~D
_~ ~ ~ uq t uq U ~3 C~ ~ Ln ~ a ~
/V H Lr~ ~ ~D ~ u ,~ N
h ~ o ~ ~ ~ ~; h ~ ~ ,1 o a~ ~ ~ uq o t~ ~ ~ o o o ~ :~ q) ~ I` ~
U Ln o o o.. ~ 4-~ al t~ 1~ o (~I O r~ ~D 00 Ln ~ C~ o ~ Ln ,~ 11~ uq u~
Vuq q~ o o o 1 0 ~
r o C a p ~_ ~ ~ a u~
r ~ U ~ C _ o H ~ O ~ a z ~L a _ ~ h ~; a ~ H :- ~r 5 Ll E~ ~ P a - u ~ v Z Z . a ~ Q ~ h ~ a : ~ -- V Q~ r ~
p Z ~ 0~ C h C r~
L~ Z ~P~ I lq Z 0 3~ 0 C
H U E-~
SUB~ JTE SHEEl~
W095/23336 ~ 8 ~ 2 3 2 PCT/CA9S/00093 We have made studies of the possible financial benefits of the present method and apparatus to swine farming, which are summarized as follows:
a. Reduced mortality rate: By allowing earlier diagnosis of diseases in a finishing barn, a reduction in mortality should result. Current average mortality is 3-4%. It should be possible to reduce this to a level of 1.5-2% by early diagnosis and treatment. This reduction of mortality rate will reduce the cost of production per pig of $2.00 to $2.50 per pig. For a 100 sow pig farm, this yields an annual saving of $4000-$5000.
b. Fewer ~poor growing" pigs: Improvements in health will improve growth rates on pig farms. It will also reduce the number of chronic ~poor growing" pigs which impair overall growth on pig farms. By reducing the number of "poor growing" pigs, the ~average days to market" number will drop. A modest improvement, such as 5 days less to market, will reduce costs by $2.50 per pig.
c. Improved days-to-market: By being able to monitor growth accurately, producers will be able to identify weak areas in pig growth, and to correct them. A modest improvement in days-to-market on average farms (5 days) will reduce cost a further $2.00-~2.50 per pig.
d. Less time looking for market-weight pigs: T h e concept of being able to monitor pig growth will allow farmers to identify pigs that are market weight. This will reduce the amount of time spent looking for market weight pigs. On a 100 sow farm, two hours per week could be saved, which would translate into a $1000 saving per year (or $1.00 per pig).
It will be obvious to one skilled in the art that modifications to the specific examples used for illustration may be made without departing from the spirit and scope of this invention.
SU~ JTE SHEET
2 ~ 8~32 units for both X and Y
coordinates; R= rastor, M M = m i l l i m e t e r s , CM= centimeters, M= meters, IN= inches, FT= feet3. Orient Ang . . Angle of Orientation: units are "degrees" of rotation from zero; zero is pointing to right of image screen '~ I 83232 woss/23336 rcTlcAs5looos3 Table VI below illustrates a typical Pen Details Report that may be printed at the end of each pen, ie. as an available report layout under operation of the preferred software. This ~ticker tape" type of report would provide - 5 immediate feedback and information to the herdsman or farm manager, eg. while the unit is proceeding through a barn.
The latter example anticipates that the camera would be on an overhead guide or track and could be selectively moved from one area for surveillance to another.
SUB~ 111 UTE SHEEI' ~ 26 --a~ -a~
,~
a ~
~ Ei -- h ~n '~ h .~ O
v ~ -- ~
r ~ 2 Y O` ~ o ., s S ~ ~ ~ x ~ _ U '1) ~ E
a~ ~ " u~ "
a 3 o ~ ,~
O ~ ~ E
'{5 0 ~ x ~' U
a~
h h ~ r~
h ~ 5~ e N
v ~
ta u _ N U N r~l ~ ~ S ' ~ V ~
P~ a' ~ 8 a~ x N ~ ~ _ a~ a~
h ~
~ ~ EE o~ ~ ~ _ h ~ _ ~- 2 ~~ r~
0 ~ ~J~ 2 e ~c~ ~ N ~ ' a) a) ~ 2 .. .. 2 ~_ 3 N N ~1 N
h ~ l~ - c _ C ~ -- N N N
0 .,1 ~ I J 1~ G -- O O O O
V U
U
a) ~
3 ~s a, ,~ J
U P~ H
a) ~ a ~ -p~ ~ h ~
SUB~ 111 ~JTE SHEET
~ I ~ 3 ~ 3 2 C~ ~ V V~ ~ ., V C~
0 ~ ~ 0 C~ o ~ 0 ~
._ . --. _ ._ , , ~ ._ , , xv~ -- Q x v~ _ E x ~n _ E
o Q ~_ ~ _ Q ~_ ~ Q ~ o, Q 0 E~ v~ E 0 E
~Y ~ 8 ~E
X X X
U U~ U
E
- _ G u~ u~ IZI u~
C ~ C~J -- ~ ~ C
e ~ _ eN _ e X N t~i r~ -- X~ i ~ -- X ~`J
Q Q Q
E ~ ~ ~ E O` ~ ~ E O`
~ - 0 C>~ 0 ~ f~ ~C~ 0 0 ~ . ~-- 0 c~ ~ v ~ v ~ c ~ v m ~ v z c~ ~n ~ ~ -- 2 Cl u~ r7 ~ -- z Cl v ~ 0 ~-- 0 C:1 ~ 1. ~ O` 0 0 0 u ~J o o -- -- u ~J O O -- --Cl N O
-- O o O O ~ -- OO O O G _ o - SUBS~ 1TE SHEE~
W095/23336 2 1 8 3 ~ 3 2 PCT/CA95/00093 Table VII below illustrates a Summary Report that may be printed at the end of the entire circuit of pens.
The livestock manager can therefore be provided with a condensed, pen-based, inventory showing total numbers of pigs: in each pen; their temperatures; and, their weights for comparison to desired shipping weights for exampl 7 SUB~ 111 UTE SHEET
WO95t23336 ~1~ 3 2 3 2 pcTlcAsslooo93 TABLE VII
Summary Report S U M M A R Y
_______ Barn-Name: Swineman, Joe Species: PORCINE
Lens: 18mm** Shipping-Weight: 105KG
File: SWINEMAN.JOE* Normal-Body-Temp: 39C
March 12, 1993 # with # at Total#
Time PenHiTemp ShipWt in Pen _______ ________ ______ ______ 9:34pm PEN2AB-2 3 4 25 9:39pm PEN2AB-1 2 5 15 9:44pm PEN2AB-4 0 2 13 9:49pm PEN2AB-3 1 0 19 SU~;~ JTE SHEEl' wos~/23336 2 1 8 3 2 3 2 PCT/CA95/00093 The preferable external interface for the processed data under the present method and apparatus is the data set of pen-based observations. This interface could also be achieved by a removable floppy diskette for transporting the data set(s) to a microcomputer for use in the database management program. It is also contemplated that the data set could be transmitted in other ways such as wireless modem or hardwired asynchronous communications line.
FIG. 2 indicates that estimating the weight of pigs under the present method has been achieved in our experiments. Plotting actual weight (scale weight - Y
axis) vs pixel value (X axis) shows a strong, useful relationship exists. Although less data is available for correlating average pixel brightness to actual pig temperature, FIG. 3 does establish that a useful relationship exists between such data and that broad estimates of an animal's temperature can be measured remotely. Our tests also indicate that thermographic remote sensing appears to be able to predict pig carcass pH, as shown in Table VIII below and in Figure 5.
/
SU8:~ 1 1 1 UTE SHER
WO 95/23336 2 1 3 ~ ~ 3 2 PCT/CA95/00093 _I
~1 o g :, n a~
~3 ~ ,~
o o~
o ~ ~ ~ o 11 ~ o o o o t H ~1 0 CO
U~ O O O O
U.
~ I U
n ~V ~
uq ~ t~ 1 11 r o O
E~ ~ ~ ~ ~ u ~ ~ ~q 3 u ~ ~ ~
~ V
H L~ V ~V
H ~q _, H~q ~ ~ w ~ ,r ~ ~I Ln ~ O_I ~ Ln Lr~ ~
~r Ln ~1 ~ ~ ~ ~
. Ln o o 1~ ~ ~ Ln o ~:~ o o o o ~V a~ ~ o o U~ O O O O r~ ~ E O O
h ~
o 11 ~ h o o a o ~ ~ Ln 1`
uq E~ 5: C~ N ~ ~ ~D
~ Z -- ~V ~D
_~ ~ ~ uq t uq U ~3 C~ ~ Ln ~ a ~
/V H Lr~ ~ ~D ~ u ,~ N
h ~ o ~ ~ ~ ~; h ~ ~ ,1 o a~ ~ ~ uq o t~ ~ ~ o o o ~ :~ q) ~ I` ~
U Ln o o o.. ~ 4-~ al t~ 1~ o (~I O r~ ~D 00 Ln ~ C~ o ~ Ln ,~ 11~ uq u~
Vuq q~ o o o 1 0 ~
r o C a p ~_ ~ ~ a u~
r ~ U ~ C _ o H ~ O ~ a z ~L a _ ~ h ~; a ~ H :- ~r 5 Ll E~ ~ P a - u ~ v Z Z . a ~ Q ~ h ~ a : ~ -- V Q~ r ~
p Z ~ 0~ C h C r~
L~ Z ~P~ I lq Z 0 3~ 0 C
H U E-~
SUB~ JTE SHEEl~
W095/23336 ~ 8 ~ 2 3 2 PCT/CA9S/00093 We have made studies of the possible financial benefits of the present method and apparatus to swine farming, which are summarized as follows:
a. Reduced mortality rate: By allowing earlier diagnosis of diseases in a finishing barn, a reduction in mortality should result. Current average mortality is 3-4%. It should be possible to reduce this to a level of 1.5-2% by early diagnosis and treatment. This reduction of mortality rate will reduce the cost of production per pig of $2.00 to $2.50 per pig. For a 100 sow pig farm, this yields an annual saving of $4000-$5000.
b. Fewer ~poor growing" pigs: Improvements in health will improve growth rates on pig farms. It will also reduce the number of chronic ~poor growing" pigs which impair overall growth on pig farms. By reducing the number of "poor growing" pigs, the ~average days to market" number will drop. A modest improvement, such as 5 days less to market, will reduce costs by $2.50 per pig.
c. Improved days-to-market: By being able to monitor growth accurately, producers will be able to identify weak areas in pig growth, and to correct them. A modest improvement in days-to-market on average farms (5 days) will reduce cost a further $2.00-~2.50 per pig.
d. Less time looking for market-weight pigs: T h e concept of being able to monitor pig growth will allow farmers to identify pigs that are market weight. This will reduce the amount of time spent looking for market weight pigs. On a 100 sow farm, two hours per week could be saved, which would translate into a $1000 saving per year (or $1.00 per pig).
It will be obvious to one skilled in the art that modifications to the specific examples used for illustration may be made without departing from the spirit and scope of this invention.
SU~ JTE SHEET
Claims (25)
1. An apparatus for remote sensing of weight information about an animal in an area and making said information available in a decipherable form to an observer, comprising:
means without human operation for remotely obtaining a thermographic image of the area comprising pixels;
means for converting said image into computer readable form, including means for converting said pixels into a number array, each number of said number array having a position corresponding to a pixel of said image and each said number having a value corresponding to a relative brightness of said pixel;
means for interpreting said form to provide said information, includlng processing means for segmenting a group of pixels associated with said animal from pixels associated with background thereby providing an animal profile of said animal and applying a unique identifier to said animal profile, said processing means further including means for counting a total number of pixels in said animal profile to determine a weight estimate of said animal: and means for displaying said information in a decipherable form.
means without human operation for remotely obtaining a thermographic image of the area comprising pixels;
means for converting said image into computer readable form, including means for converting said pixels into a number array, each number of said number array having a position corresponding to a pixel of said image and each said number having a value corresponding to a relative brightness of said pixel;
means for interpreting said form to provide said information, includlng processing means for segmenting a group of pixels associated with said animal from pixels associated with background thereby providing an animal profile of said animal and applying a unique identifier to said animal profile, said processing means further including means for counting a total number of pixels in said animal profile to determine a weight estimate of said animal: and means for displaying said information in a decipherable form.
2. An apparatus as defined in claim 1, wherein said processing means includes means for segmenting a single group of pixels associated with a plurality of animals in said area and determining an approximation of a number of animals associated with said single group of pixels.
3. An apparatus as defined in claim 2, wherein said processing means includes means for counting said unique identifier and said approximation of a number of animals associated with said single group of pixels, to determine an estimate of a number of animals in the area.
4. An apparatus as defined in claim 1, wherein the means for remotely obtaining a thermographic image of the area is a thermographic imaging camera.
5. An apparatus as defined in claim 1, wherein said means for converting said image into computer readable form includes a video-digitizer circuit board.
6. An apparatus as defined in claim 4, wherein said camera is mountable on a means for moving said camera over said area and at least one other area for selective remote sensing of an animal in said area or said at least one other area.
7. An apparatus as defined in claim 6, wherein said means for moving the camera includes a track, extending from over said area to said at least one other area, and on which said camera is slidably mounted.
8. An apparatus as defined in claim 4, wherein said camera is fixable to view only said area.
9. An apparatus as defined in claim 1, wherein said processing means includes means for determining a centre of said animal from locations of all pixels in said animal profile and then determining an approximate position of said animal in the area.
10. An apparatus as defined in claim 1, wherein said processing means includes means for determining an approximate absolute or relative temperature of said animal based on average brightness of pixels in said animal profile.
11. An apparatus as defined in claim 10, wherein said processing means also includes means for determining from said approximate temperature prior to slaughter of said animal, a mathematically correlated prediction of approximate carcass pH twenty-four hours after slaughter of said animal.
12. An apparatus as defined in claim 10, wherein said processing means also includes means for determining from said approximate temperature prior to slaughter of said animal, a mathematically correlated prediction of approximate post-slaughter visual Pale Soft Exudative meat score.
13. An apparatus as defined in claim 1, wherein said processing means includes means for determining an approximate absolute or relative temperature of said background based on average brightness of said pixels associated with background.
14. A method of remotely sensing weight information about an animal in an area, which comprises:
obtaining, without human operation, a thermographic image of the area comprising pixels;
converting said image into computer readable form;
interpreting said form to provide said information, including segmenting a group of pixels associated with said animal from pixels associated with background thereby providing an animal profile of said animal and applying a unique identifier to said animal profile, and counting a total number of pixels in said animal profile to determine a weight estimate of said animal; and displaying said information in a decipherable form.
obtaining, without human operation, a thermographic image of the area comprising pixels;
converting said image into computer readable form;
interpreting said form to provide said information, including segmenting a group of pixels associated with said animal from pixels associated with background thereby providing an animal profile of said animal and applying a unique identifier to said animal profile, and counting a total number of pixels in said animal profile to determine a weight estimate of said animal; and displaying said information in a decipherable form.
15. A method as defined in claim 14, wherein said interpreting includes segmenting a single group of pixels associated with a plurality of animals in said area and determining an approximation of a number of animals associated with said single group of pixels.
16. A method as defined in claim 15, wherein said interpreting also includes counting said unique identifier and said approximation of a number of animals associated with said single group of pixels, to determine an estimate of a number of animals in the area.
17. A method as defined in claim 14, wherein said interpreting also includes determining a centre of said animal from locations of all pixels in said animal profile and then determining an approximate position of said animal in the area.
18. A method as defined in claim 14, wherein said interpreting also includes determining an approximate absolute or relative temperature of said animal based on average brightness of pixels in said animal profile.
19. A method as defined in claim 14, wherein said interpreting includes determining an approximate absolute or relative temperature of said background based on average brightness of said pixels associated with background.
20. A method as defined in claim 18, wherein said interpreting also includes determining from said approximate temperature prior to slaughter of said animal, a mathematically correlated prediction of approximate carcass pH twenty-four hours after slaughter of said animal.
21. A method as defined in claim 18, wherein said interpreting also includes determining from said approximate temperature prior to slaughter of said animal, a mathematically correlated prediction of approximate post-slaughter visual Pale Soft Exudative meat score of said animal.
22. An apparatus for remote sensing of information about a number of animals in an area and making said information available in a decipherable form to an observer, comprising:
means without human operation for remotely obtaining a thermographic image of the area, said image comprising pixels;
means for converting said image into computer readable form, including means for converting said pixels into a number array, each number of said number array having a position corresponding to a pixel of said image and each said number having a value corresponding to a relative brightness of said pixel;
means for interpreting said form to provide said information, including processing means for segmenting a group of pixels associated with an animal from pixels associated with background thereby providing an animal profile of said animal and applying a unique identifier to said animal profile, and further including processing means for segmenting a single group of pixels associated with a plurality of animals in said area and determining an approximation of a number of animals associated with said single group of pixels, and further including means for counting said unique identifier and said approximation of a number of animals associated with said single group of pixels, to determine an estimate of a number of animals in the area; and means for displaying said information in a decipherable form.
means without human operation for remotely obtaining a thermographic image of the area, said image comprising pixels;
means for converting said image into computer readable form, including means for converting said pixels into a number array, each number of said number array having a position corresponding to a pixel of said image and each said number having a value corresponding to a relative brightness of said pixel;
means for interpreting said form to provide said information, including processing means for segmenting a group of pixels associated with an animal from pixels associated with background thereby providing an animal profile of said animal and applying a unique identifier to said animal profile, and further including processing means for segmenting a single group of pixels associated with a plurality of animals in said area and determining an approximation of a number of animals associated with said single group of pixels, and further including means for counting said unique identifier and said approximation of a number of animals associated with said single group of pixels, to determine an estimate of a number of animals in the area; and means for displaying said information in a decipherable form.
23. An apparatus as defined in claim 22, wherein said means for interpreting includes means for determining an approximate absolute or relative temperature of said animal based on average brightness of pixels in said animal profile.
24. A method of remotely sensing information about a number of animals in an area, which comprises:
obtaining, without human operation, a thermographic image of the area comprising pixels;
converting said image into computer readable form;
interpreting said form to provide said information, including: segmenting a group of pixels associated with an animal from pixels associated with background thereby providing an animal profile of said animal and applying a unique identifier to said animal profile; segmenting a single group of pixels associated with a plurality of animals in said area and determining an approximation of a number of animals associated with said single group of pixels; and counting said unique identifier and said approximation of a number of animals associated with said single group of pixels to determine an estimate of a number of animals in the area; and displaying said information in a decipherable form.
obtaining, without human operation, a thermographic image of the area comprising pixels;
converting said image into computer readable form;
interpreting said form to provide said information, including: segmenting a group of pixels associated with an animal from pixels associated with background thereby providing an animal profile of said animal and applying a unique identifier to said animal profile; segmenting a single group of pixels associated with a plurality of animals in said area and determining an approximation of a number of animals associated with said single group of pixels; and counting said unique identifier and said approximation of a number of animals associated with said single group of pixels to determine an estimate of a number of animals in the area; and displaying said information in a decipherable form.
25. A method as defined in claim 24, wherein said interpreting also includes determining an approximate absolute or relative temperature of said animal based on average brightness of pixels in said animal profile.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US08/201,168 | 1994-02-24 | ||
| US08/201,168 US5474085A (en) | 1994-02-24 | 1994-02-24 | Remote thermographic sensing of livestock |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA2183232A1 true CA2183232A1 (en) | 1995-08-31 |
Family
ID=22744746
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA002183232A Abandoned CA2183232A1 (en) | 1994-02-24 | 1995-02-24 | Remote thermographic sensing of livestock |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US5474085A (en) |
| AU (1) | AU1703895A (en) |
| CA (1) | CA2183232A1 (en) |
| WO (1) | WO1995023336A1 (en) |
Families Citing this family (72)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5595444A (en) * | 1993-07-02 | 1997-01-21 | Her Majesty The Queen In Right Of Canada, As Represented By The Department Of Agriculture And Agri-Food Canada | Method for detecting poor meat quality in groups of live animals |
| US5979359A (en) * | 1995-10-27 | 1999-11-09 | Alfa Laval Agri Ab | Analysis of color tone in images for use in animal breeding |
| BR9711229A (en) * | 1996-08-23 | 2001-11-06 | United Kingdom Government | Process and apparatus for using image analysis to determine meat and carcass characteristics |
| US6123451A (en) * | 1997-03-17 | 2000-09-26 | Her Majesty The Queen In Right Of Canada, As Represented By The Administer For The Department Of Agiculture And Agri-Food (Afcc) | Process for determining a tissue composition characteristic of an animal |
| US6023637A (en) * | 1997-03-31 | 2000-02-08 | Liu; Zhong Qi | Method and apparatus for thermal radiation imaging |
| NL1006296C2 (en) * | 1997-06-12 | 1998-12-15 | Nedap Nv | Detection of oestrum in e.g. dairy herd |
| US5887547A (en) * | 1997-07-03 | 1999-03-30 | Enviromentally Correct Concepts, Inc. | Method for measuring and quantifying amounts of carbon from certain greenhouse gases sequestered in grassy and herbaceous plants above and below the soil surface |
| BR9908065A (en) | 1998-02-20 | 2001-11-13 | Univ Colorado State Res Found | Meat color imaging system for pleasant taste and yield prognosis |
| US6375612B1 (en) | 1998-03-24 | 2002-04-23 | P. Timothy Guichon | Method and system for monitoring animals |
| WO2000049400A1 (en) * | 1999-02-18 | 2000-08-24 | Colorado State University Research Foundation | Meat imaging system for palatability and yield prediction |
| AU4020200A (en) * | 1999-03-22 | 2000-10-09 | Emerge Interactive, Inc. | Early detection of inflammation and infection using infrared thermography |
| US7277744B2 (en) * | 1999-03-22 | 2007-10-02 | Schaefer Allan L | Early detection of inflammation and infection using infrared thermography |
| NZ514340A (en) * | 1999-03-22 | 2004-01-30 | Allan L Schaefer | Early detection of inflammation and infection using infrared thermography |
| DE19918100B4 (en) * | 1999-04-22 | 2009-05-20 | Niklaus, Helmut, Dr. | Device and method for contactless weight determination of live animals, especially of pigs in their housing environment, preferably in their bays |
| NL1012137C2 (en) * | 1999-05-25 | 2000-11-28 | Lely Res Holding | Unmanned vehicle that can be used in a stable or a meadow. |
| NL1012138C2 (en) * | 1999-05-25 | 2000-11-28 | Lely Res Holding | Climate cart. |
| ATE301928T1 (en) * | 1999-08-23 | 2005-09-15 | Gray David Reynolds | ACTIVITY MEASUREMENT AND OBSERVATION OF ANIMALS AND PLANTS |
| US7724925B2 (en) * | 1999-12-02 | 2010-05-25 | Thermal Wave Imaging, Inc. | System for generating thermographic images using thermographic signal reconstruction |
| US6681717B2 (en) * | 2000-12-15 | 2004-01-27 | Can Technologies, Inc. | Computer system for determining a customized animal feed |
| NL1017354C2 (en) * | 2001-02-13 | 2002-08-14 | Lely Entpr Ag | Device and method for milking an animal, device for monitoring an animal. |
| US20050153359A1 (en) * | 2002-03-11 | 2005-07-14 | Schaefer Allan L. | Method for the evaluation of velvet antler |
| US6974373B2 (en) * | 2002-08-02 | 2005-12-13 | Geissler Technologies, Llc | Apparatus and methods for the volumetric and dimensional measurement of livestock |
| US7039220B2 (en) * | 2002-08-14 | 2006-05-02 | C-Scan, L.L.P. | Methods and apparatus for the dimensional measurement of livestock using a single camera |
| US6805074B2 (en) | 2002-12-23 | 2004-10-19 | Cargill, Incorporated | System and method of customizing an animal feed based on heat increment |
| US20040154550A1 (en) * | 2003-01-08 | 2004-08-12 | Mcquilkin Gary L. | Methods and apparatus for a remote, noninvasive technique to detect chronic wasting disease (CWD) and similar diseases in live subjects |
| AU2003901694A0 (en) * | 2003-04-10 | 2003-05-01 | Singapore Technologies Electronics Limited | Method and apparatus for measuring body temperature |
| US20040236191A1 (en) * | 2003-05-19 | 2004-11-25 | Poliska Steven A. | System and method for identifying and labeling livestock products, and managing data associated with those products |
| NZ543950A (en) | 2003-05-30 | 2007-01-26 | Harry J S Harrison | The use of infrared thermography in live animals to predict growth efficiency |
| CN1311783C (en) * | 2004-03-10 | 2007-04-25 | 刘忠齐 | Method for evaluating effect of edjustment of physiological and mental state |
| EP1579804A1 (en) * | 2004-03-22 | 2005-09-28 | Infradan ApS | Detecting a physical condition of health of an animal |
| US20080189085A1 (en) * | 2004-07-29 | 2008-08-07 | Can Technologies, Inc. | System and method for optimizing animal production |
| US20060036419A1 (en) * | 2004-07-29 | 2006-02-16 | Can Technologies, Inc. | System and method for animal production optimization |
| CA2596837A1 (en) * | 2004-11-29 | 2006-06-01 | Cargill, Incorporated | Animal management system |
| GB0716333D0 (en) * | 2007-08-22 | 2007-10-03 | White Spark Holdings Ltd | Method and apparatus for the automatic grading of condition of livestock |
| DE202008007880U1 (en) * | 2008-06-13 | 2009-10-29 | Big Dutchman International Gmbh | poultry scale |
| US8297231B2 (en) * | 2009-02-03 | 2012-10-30 | Faunus Ltd. | System and methods for health monitoring of anonymous animals in livestock groups |
| US20100198024A1 (en) * | 2009-02-03 | 2010-08-05 | Ron Elazari-Volcani | Vitality meter for health monitoring of anonymous animals in livestock groups |
| WO2010127023A1 (en) * | 2009-05-01 | 2010-11-04 | Spicola Tool, Llc | Remote contactless stereoscopic mass estimation system |
| AU2011218640B2 (en) | 2010-12-09 | 2014-12-11 | Smart Farm Technologies Limited | Detection apparatus |
| SE535435C2 (en) * | 2010-12-15 | 2012-08-07 | Agricam Ab | A system and method for controlling an automatic milking system |
| US8755570B2 (en) * | 2011-04-27 | 2014-06-17 | Steve Gomas | Apparatus and method for estimation of livestock weight |
| WO2012149654A1 (en) * | 2011-05-02 | 2012-11-08 | Agridigit Inc | Evaluation of animal products based on customized models |
| US8787621B2 (en) * | 2012-06-04 | 2014-07-22 | Clicrweight, LLC | Methods and systems for determining and displaying animal metrics |
| US9167800B2 (en) | 2012-06-04 | 2015-10-27 | Clicrweight, LLC | Systems for determining animal metrics and related devices and methods |
| WO2014083433A2 (en) * | 2012-12-02 | 2014-06-05 | Agricam Ab | Systems and methods for predicting the outcome of a state of a subject |
| US20140267768A1 (en) * | 2013-03-15 | 2014-09-18 | Mu Optics, Llc | Thermographic Camera Accessory for Personal Electronics |
| NL2011952C2 (en) * | 2013-12-12 | 2015-06-15 | Lely Patent Nv | Method and device to monitor growth of an animal, in particular a calf. |
| US9961883B2 (en) * | 2014-06-16 | 2018-05-08 | Allan Schaefer | Rapid and automatic determination of metabolic efficiency in livestock |
| US9955672B2 (en) | 2014-06-16 | 2018-05-01 | Nigel Cook | Infrared thermography and behaviour information for identification of biologically important states in animals |
| US10956856B2 (en) | 2015-01-23 | 2021-03-23 | Samsung Electronics Co., Ltd. | Object recognition for a storage structure |
| CA2977911C (en) | 2015-02-27 | 2023-11-14 | Ingenera Sa | Improved method and apparatus for the determination of the body condition score, body weight and state of fertility |
| US9745060B2 (en) | 2015-07-17 | 2017-08-29 | Topcon Positioning Systems, Inc. | Agricultural crop analysis drone |
| US10231441B2 (en) | 2015-09-24 | 2019-03-19 | Digi-Star, Llc | Agricultural drone for use in livestock feeding |
| US10321663B2 (en) | 2015-09-24 | 2019-06-18 | Digi-Star, Llc | Agricultural drone for use in livestock monitoring |
| DE102016107208A1 (en) * | 2016-04-19 | 2017-10-19 | Rheinmetall Defence Electronics Gmbh | Method, device and system for the automatic detection of inflamed parts of the body of a living being |
| WO2018015519A1 (en) * | 2016-07-20 | 2018-01-25 | Farm Robotics And Automation Sl | Robot assisted surveillance of livestock |
| JP7002066B2 (en) * | 2017-07-31 | 2022-01-20 | 株式会社コンピューター総合研究所 | Information processing systems, information processing equipment, and programs |
| WO2019058752A1 (en) * | 2017-09-22 | 2019-03-28 | パナソニックＩｐマネジメント株式会社 | Livestock information management system, livestock barn, livestock information management program, and livestock information management method |
| DE202017106101U1 (en) * | 2017-10-09 | 2019-02-18 | Big Dutchman International Gmbh | Device for finding objects in an animal house |
| ES2921128T3 (en) | 2018-02-26 | 2022-08-18 | Touchless Animal Metrics Sl | Procedure and device for the characterization of living specimens remotely |
| US10362769B1 (en) * | 2018-03-23 | 2019-07-30 | International Business Machines Corporation | System and method for detection of disease breakouts |
| KR102030768B1 (en) * | 2018-05-08 | 2019-10-10 | 숭실대학교산학협력단 | Poultry weight measuring method using image, recording medium and device for performing the method |
| WO2020025320A1 (en) | 2018-07-31 | 2020-02-06 | Signify Holding B.V. | Controller for detecting animals with physiological conditions |
| PT3866587T (en) | 2018-10-17 | 2023-04-26 | Group Ro Main Inc | Monitoring livestock |
| US20220000077A1 (en) * | 2018-11-14 | 2022-01-06 | The State Of Israel, Ministry Of Agriculture & Rural Development, Agricultural Research Organization | System for measuring body temperature in poultry |
| WO2020110616A1 (en) * | 2018-11-29 | 2020-06-04 | パナソニックＩｐマネジメント株式会社 | Poultry raising system, poultry raising method, and program |
| US10863724B2 (en) * | 2018-12-11 | 2020-12-15 | Animal Health Analytics, Inc | System and method for tracking and scoring animal health and meat quality |
| CN111418509A (en) * | 2020-01-16 | 2020-07-17 | 秒针信息技术有限公司 | Method and device for feeding livestock, storage medium and electronic device |
| CN111339912B (en) * | 2020-02-24 | 2023-04-18 | 中国科学院植物研究所 | Method and system for recognizing cattle and sheep based on remote sensing image |
| BR112022021057A2 (en) * | 2020-04-17 | 2022-12-06 | Schaefer Allan | SYSTEM AND METHOD IMPLEMENTED BY COMPUTER |
| CN111369560B (en) * | 2020-04-26 | 2023-04-28 | 成都大熊猫繁育研究基地 | Method for rapidly measuring body temperature of containment panda |
| IT202100011255A1 (en) * | 2021-05-03 | 2022-11-03 | Vem Sistemi S P A | Live animal growth monitoring system and monitoring procedure |
Family Cites Families (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA854384A (en) * | 1970-10-20 | Shimotsuma Teruo | Apparatus for displaying a temperature distribution pattern | |
| US3738320A (en) * | 1970-08-10 | 1973-06-12 | Ross Holm Division Holm Tracto | Automatic milking barn |
| US4349843A (en) * | 1978-06-26 | 1982-09-14 | Flir Systems, Inc. | Television compatible thermal imaging system |
| BE874865A (en) * | 1979-03-15 | 1979-07-02 | Schlager Kenneth J | THERMOGRAPHIC DEVICE FOR THE PHYSICAL EXAMINATION OF PATIENTS |
| DE3176016D1 (en) * | 1980-12-30 | 1987-04-23 | Ibm | System for remotely displaying and sensing information using shadow parallax |
| US4691712A (en) * | 1983-07-20 | 1987-09-08 | American Thermometer Co., Inc. | Device for detecting, measuring, and recording body thermal emissivity |
| US5056525A (en) * | 1984-03-21 | 1991-10-15 | Hafezi Fred F | Process for diagnosing spinal conditions |
| US4679077A (en) * | 1984-11-10 | 1987-07-07 | Matsushita Electric Works, Ltd. | Visual Image sensor system |
| US4617876A (en) * | 1984-11-28 | 1986-10-21 | Hayes Norman J | Animal identification and control system |
| GB2183878B (en) * | 1985-10-11 | 1989-09-20 | Matsushita Electric Works Ltd | Abnormality supervising system |
| US4917117A (en) * | 1986-09-04 | 1990-04-17 | Ciba-Geigy Corporation | Process and apparatus for the recording and/or evaluation of the movement behavior of experimental animals |
| US4917048A (en) * | 1987-09-15 | 1990-04-17 | John Beattie | Automatic trapping system |
| DK552887A (en) * | 1987-10-22 | 1989-04-23 | Slagteriernes Forskningsinst | PROCEDURES FOR TREATING Slaughter pigs at slaughterhouses and facilities for the preparation of slaughter pigs |
| US4866276A (en) * | 1987-12-29 | 1989-09-12 | The Boeing Company | Method and apparatus for nondestructive analysis of subsurface features of material |
| JP2552728B2 (en) * | 1989-05-31 | 1996-11-13 | 富士通株式会社 | Infrared surveillance system |
| JP2633694B2 (en) * | 1989-08-25 | 1997-07-23 | フジテック 株式会社 | Person detection device |
| JPH0690969A (en) * | 1991-01-11 | 1994-04-05 | Tadamasa Aoyama | Method for measuring living body such as horse |
| US5163094A (en) * | 1991-03-20 | 1992-11-10 | Francine J. Prokoski | Method for identifying individuals from analysis of elemental shapes derived from biosensor data |
| JPH04373371A (en) * | 1991-06-24 | 1992-12-25 | Matsushita Electric Ind Co Ltd | Video camera system with thermal picture detecting means |
| JP2682302B2 (en) * | 1991-10-21 | 1997-11-26 | 株式会社村田製作所 | Infrared detector |
| NO923094L (en) * | 1992-08-07 | 1994-02-08 | Sinvent As | Procedure for determining the nature and distribution of tissue types in living organisms, especially food content in eg crabs |
-
1994
- 1994-02-24 US US08/201,168 patent/US5474085A/en not_active Expired - Fee Related
-
1995
- 1995-02-24 WO PCT/CA1995/000093 patent/WO1995023336A1/en active Application Filing
- 1995-02-24 CA CA002183232A patent/CA2183232A1/en not_active Abandoned
- 1995-02-24 AU AU17038/95A patent/AU1703895A/en not_active Abandoned
Also Published As
| Publication number | Publication date |
|---|---|
| AU1703895A (en) | 1995-09-11 |
| WO1995023336A1 (en) | 1995-08-31 |
| US5474085A (en) | 1995-12-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5474085A (en) | Remote thermographic sensing of livestock | |
| Schofield et al. | Monitoring pig growth using a prototype imaging system | |
| Gómez et al. | A systematic review on validated precision livestock farming technologies for pig production and its potential to assess animal welfare | |
| Halachmi et al. | Automatic assessment of dairy cattle body condition score using thermal imaging | |
| Maselyne et al. | Validation of a High Frequency Radio Frequency Identification (HF RFID) system for registering feeding patterns of growing-finishing pigs | |
| US5595444A (en) | Method for detecting poor meat quality in groups of live animals | |
| EP1060391B1 (en) | Meat color imaging system for palatability and yield prediction | |
| US6375612B1 (en) | Method and system for monitoring animals | |
| Stajnko et al. | Estimation of bull live weight through thermographically measured body dimensions | |
| CN104188635B (en) | Livestock living body measurement equipment and method | |
| US9955672B2 (en) | Infrared thermography and behaviour information for identification of biologically important states in animals | |
| AU733733B2 (en) | Process for determining a tissue composition characteristic of an animal | |
| Jackson et al. | Changes in feeding behavior patterns and dry matter intake before clinical symptoms associated with bovine respiratory disease in growing bulls | |
| WO2014145552A1 (en) | Automated monitoring of ruminant health and breeding parameters | |
| KR20210031091A (en) | System and Method for Monitoring Activity and Forecasting Healthcare of Livestock | |
| CN114155216A (en) | Pig temperature detection method and device | |
| CN116019023A (en) | Electronic information management system and management method | |
| Imaz et al. | Using automated in-paddock weighing to evaluate the impact of intervals between liveweight measures on growth rate calculations in grazing beef cattle | |
| Hansen et al. | Non-intrusive automated measurement of dairy cow body condition using 3D video | |
| EP3769036A1 (en) | Method and system for extraction of statistical sample of moving objects | |
| Singh et al. | Precision dairy farming: the next dairy marvel. | |
| CN113068657B (en) | Intelligent efficient pig raising method and system | |
| CA2854344C (en) | Infrared thermography and behaviour information for identification of biolically important states in animals | |
| EP3606349A1 (en) | Method and system for classifying animal carcass | |
| EP4104676A1 (en) | Method and system for monitoring estrus and ovulation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request | ||
| FZDE | Discontinued |