CN201935873U - Online image detection system for bottle cap - Google Patents

Abstract

The utility model belongs to the technical field of machine vision and image processing, relating to an online image detection system for a bottle cap. The online image detection system comprises a conveyor belt for conveying the bottle cap, an optical sensor, a camera, a light source, an image acquisition card and a computer; the camera is fixed over the conveyor belt; the light source is arranged above the conveyor belt; the optical sensor is arranged beside the conveyor belt under the camera for detecting whether the bottle cap exists under the camera; the camera is triggered by the detection signal of the optical sensor to collect the image of the bottle cap positioned under the camera; and the collected image of the bottle cap is transmitted into the computer through the image acquisition card. In the utility model, the light source is reasonably distributed, the camera is triggered through the optical sensor to collect the image online, the image feature extraction and high-speed automatic detection are realized by the computer, and the technical production quality of the product package can be improved.

Description

The online image detection system of bottle cap

Technical field

The utility model belongs to machine vision and technical field of image processing, relates to the online image detection system of a kind of bottle cap.

Background technology

Domestic many manufacturers recognize that more and more quality and throughput rate are most crucial factor this point forever now, at the industry spot machine vision technique more and more in order to monitoring industrial processes and testing product quality, thereby guarantee the performance and the presentation quality of product.It all is to finish by manpower that the product of present domestic most production lines detects, not only can increase the human input of company, also caused inefficiency, therefore, with machine vision technique finish the defects detection of product and classification, classification has great demand, also is the field that present whole vision industry is relatively made earnest efforts simultaneously.At present, vinyl cover, aluminium lid and so on bottle cap can be produced on industrial production line fast, and the defective that may exist bottle cap realizes high efficiency detection and rejecting, to adapt to quick Production Line efficient, has necessity.

The utility model content

The purpose of this utility model provides a kind of can realize the online detection of bottle cap, avoids packing of product technology to produce the bottle cap online detection instrument of quality problems.The scheme that its technical matters that solves the utility model adopts is as follows:

The online image detection system of a kind of bottle cap, comprise the travelling belt, optical sensor, camera, light source, image pick-up card, the computing machine that are used to transmit bottle cap, camera be fixed on travelling belt directly over, light source places the oblique upper of travelling belt, optical sensor places by the travelling belt under the camera, whether there is bottle cap under being used to detect camera, trigger the image that camera collection is positioned at the bottle cap on the travelling belt under it by the detection signal of optical sensor, the bottle cap image of being gathered is admitted in the computing machine through image pick-up card.

Preferably, described optical sensor is the infrared induction photoelectric sensor.

The bottle cap on-line detecting system that the utility model provides, source layout is reasonable, triggers camera online acquisition image by optical sensor, by computer realization image characteristics extraction and High-Speed Automatic detection, can improve packing of product technology and produce quality.

Description of drawings

Fig. 1: the utility model hardware system outward appearance is formed synoptic diagram;

Fig. 2: the utility model software flow pattern;

Fig. 3: the utility model defects detection subroutine flow chart;

Fig. 4: the stretching schematic flow sheet of annulus of the present utility model.

Among Fig. 1: 1 travelling belt, 2 bottle caps, 3 optical sensors, 4 sensor signal lines, 5 cameras, 6 light sources, 7 transmission lines, 8 image pick-up cards, 9 computing machines.

Embodiment

Referring to Fig. 1, bottle cap high-speed detection system of the present utility model is made up of transfer equipment, camera, special light source, transmission line, image pick-up card, computing machine.Special transmission equipment comprises travelling belt and optical sensor.The camera of video camera can be the line array CCD camera, is installed in directly over the travelling belt, triggers by optical sensor signals and gathers lid mouth bottle cap image up in real time.Special light source is angled to the bottle cap oblique illumination, thereby makes the bottle cap edge be rendered as the annulus of certain width in images acquired, so that detect the bottle cap breach.Optical sensor is the infrared induction photoelectric sensor, is fixed in by the transmission line of video camera below.Image pick-up card can be based on the outer plug-in card that microcomputer pci bus structure has the image front-end processing, with more complete real time image collection disposal system of existing resource formation of microcomputer.Because amount of image information is big, the processing time is long, described computing machine need have higher dominant frequency and stronger arithmetic capability.Camera acquisition realtime graphic, size are 800 * 600, by transmission line and image pick-up card deliver to Computer Storage be the BMP form and and detect in real time.

The utility model is made up of hardware components and software section; Hardware components adopts computing machine as handling and control center; Software section is made up of a plurality of subroutines such as image pre-service, edge extraction, deformation detection, defects detection.

Comprise referring to this software section of Fig. 2: 1) image pre-service subroutine, 2) edge extracts subroutine, 3) deformation detection subroutine, 4) the defects detection subroutine.

1) described image pre-service subroutine comprises: image gray processing, removal noise, grey level stretching, binaryzation and morphologic filtering.

Image gray processing, the image that camera is taken in real time is 16 bitmaps (RGB565 form) data, handles for the ease of follow-up rapid image, needs view data is changed, and makes coloured image become 256 grades of gray-scale maps.

Remove noise, inevitably contain noise in the image, adopt medium filtering that image is carried out pre-service thereupon.

Grey level stretching in order to strengthen the contrast of background area and character zone, is carried out grey level stretching to image.

Binaryzation is carried out binary conversion treatment to gray level image, adopts the method for maximum between-cluster variance and infima species internal variance ratio, and self-adaptation is calculated gray threshold, thinks the target area less than the zone of this threshold value, greater than the background area of thinking in the zone of this threshold value.

Morphologic filtering carries out morphologic filtering to bianry image and handles, the synthetic operation that adopts expansion, burn into ON operation and closed operation to combine.

2) edge extracts subroutine, bottle cap mouth edge is rendered as the annulus of certain width in the image after the binaryzation, select template extraction edge doughnut picture by the zone that three concentric circless of drag and drop constitute, three radius of a circles can independently be set, and justify called after cylindrical, middle circle and interior circles respectively according to order from outside to inside with three.Middle circle is used to estimate that bottle cap radius, inside and outside circle are used to define the edge annulus of bottle cap, the interior edge of wherein interior circle coupling bottle cap, cylindrical coupling bottle cap outer.

3) deformation detection subroutine is calculated annulus outer circularity as shown in Equation 1, and compares with the circularity threshold value of presetting, if result of calculation is lower than the circularity threshold value, then bottle cap is judged as distortion, produces simultaneously and rejects signal.

$C=\frac{P^2}{A}---\left(1\right)$

Wherein, P is a girth, and A is an area.

4) defects detection subroutine detects bottle cap and whether has defectives such as burr, burr and breach, comprising: calculation of parameter, annulus are stretching, pre-service and width measure, referring to Fig. 3.

Calculation of parameter: calculate the stretching projective transformation of edge annulus desired position parameter, i.e. sine and cosine value, and deposit corresponding array in.As shown in Figure 4, if the pixel (xx that certain in the edge annulus is to be stretched, yy) with the distance of x axle positive axis with respect to the center of circle (x0, y0) angle is a, sine here and cosine value are exactly sine and the cosine value that refers to angle a, be used for calculating the projective transformation stretch-out view former figure be in the original edge annulus respective coordinates (x y) obtains pixel value, if the cavity point further adopts its pixel value of interpolation calculation.

Annulus is stretching, select cylindrical in the template as benchmark with the zone, promptly with cylindrical as projective transformation after the top margin of rectangular area, the inside and outside circle semidiameter as projective transformation after the width of rectangular area, the edge annulus is projected to aforementioned rectangular area along the incision of right side horizontal line, and wherein the pixel value of cavity point adopts the bilinear interpolation algorithm computation.

Pre-service looks like to carry out binary conversion treatment according to the threshold value of binaryzation in the image pre-service to stretching doughnut, and carries out morphologic filtering, eliminates isolated point.

Width measure obtains the minimum and maximum width value of annulus and calculates the mean breadth of annulus by statistics, if minimum and maximum width value surpasses the mean breadth certain proportion, thinks that then there are defectives such as burr, burr or breach in bottle cap, produces simultaneously and rejects signal.

Below introduce preferred embodiment of the present utility model, this part only is to illustrate of the present utility model, but not to the restriction of the utility model and application or purposes.According to other embodiment that the utility model draws, belong to technological innovation scope of the present utility model too.There is the setting of related parameter also not show to have only example value to use in the scheme.

The course of work of the present utility model: system powers on, and video camera is gathered the bottle cap image continuously in real time, and arrives microcomputer by transmission line; The image pretreatment module to input picture strengthen, filtering and binary conversion treatment; Select module to extract bottle cap edge doughnut picture according to the zone that the concentric circles group constitutes; Calculate the circularity of annulus outer, and with default circularity threshold ratio, be made as 0.8 as the circularity threshold value, be distortion if result of calculation, is then judged bottle cap less than predetermined threshold value, produce and reject signal, otherwise enter the defects detection subroutine; Defect inspection process is at first calculated the stretching projective transformation of annulus required sine and cosine value, again by projective transformation the gray scale doughnut as stretching, by binaryzation and filtering stretching doughnut is looked like to handle then, add up at last annulus minimum and maximum width and with mean breadth relatively, if minimum and maximum width surpasses 20% of mean breadth, judge that then there is defective in bottle cap, produce and reject signal.

Claims (2)

1. online image detection system of bottle cap, comprise the travelling belt that is used to transmit bottle cap, optical sensor, camera, light source, image pick-up card, computing machine, it is characterized in that, camera be fixed on travelling belt directly over, light source places the oblique upper of travelling belt, optical sensor places by the travelling belt under the camera, whether there is bottle cap under being used to detect camera, trigger the image that camera collection is positioned at the bottle cap on the travelling belt under it by the detection signal of optical sensor, the bottle cap image of being gathered is admitted in the computing machine through image pick-up card.

2. the online image detection system of bottle cap according to claim 1, described optical sensor is the infrared induction photoelectric sensor.

Images