WO2007078912A1 - Method and apparatus providing color interpolation in color filter arrays using edge detection and correction terms - Google Patents
Method and apparatus providing color interpolation in color filter arrays using edge detection and correction terms Download PDFInfo
- Publication number
- WO2007078912A1 WO2007078912A1 PCT/US2006/048200 US2006048200W WO2007078912A1 WO 2007078912 A1 WO2007078912 A1 WO 2007078912A1 US 2006048200 W US2006048200 W US 2006048200W WO 2007078912 A1 WO2007078912 A1 WO 2007078912A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- pixel
- values
- color
- selected pixel
- edge
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000003708 edge detection Methods 0.000 title claims abstract description 8
- 238000012937 correction Methods 0.000 title claims description 11
- 238000003491 array Methods 0.000 title description 3
- 239000003086 colorant Substances 0.000 claims abstract description 23
- 238000003384 imaging method Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 12
- 238000001228 spectrum Methods 0.000 claims 2
- 102220103881 rs201490575 Human genes 0.000 claims 1
- 102220124522 rs746215581 Human genes 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 238000001465 metallisation Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 239000012212 insulator Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002161 passivation Methods 0.000 description 1
- 239000006187 pill Substances 0.000 description 1
- 102220124597 rs374821619 Human genes 0.000 description 1
- 102220094200 rs876659020 Human genes 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4015—Demosaicing, e.g. colour filter array [CFA], Bayer pattern
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/10—Cameras or camera modules comprising electronic image sensors; Control thereof for generating image signals from different wavelengths
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L27/00—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
- H01L27/14—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation
- H01L27/144—Devices controlled by radiation
- H01L27/146—Imager structures
- H01L27/14601—Structural or functional details thereof
- H01L27/1462—Coatings
- H01L27/14621—Colour filter arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/80—Camera processing pipelines; Components thereof
- H04N23/84—Camera processing pipelines; Components thereof for processing colour signals
- H04N23/843—Demosaicing, e.g. interpolating colour pixel values
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N25/00—Circuitry of solid-state image sensors [SSIS]; Control thereof
- H04N25/10—Circuitry of solid-state image sensors [SSIS]; Control thereof for transforming different wavelengths into image signals
- H04N25/11—Arrangement of colour filter arrays [CFA]; Filter mosaics
- H04N25/13—Arrangement of colour filter arrays [CFA]; Filter mosaics characterised by the spectral characteristics of the filter elements
- H04N25/134—Arrangement of colour filter arrays [CFA]; Filter mosaics characterised by the spectral characteristics of the filter elements based on three different wavelength filter elements
-
- H—ELECTRICITY
- H01—ELECTRIC ELEMENTS
- H01L—SEMICONDUCTOR DEVICES NOT COVERED BY CLASS H10
- H01L27/00—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate
- H01L27/14—Devices consisting of a plurality of semiconductor or other solid-state components formed in or on a common substrate including semiconductor components sensitive to infrared radiation, light, electromagnetic radiation of shorter wavelength or corpuscular radiation and specially adapted either for the conversion of the energy of such radiation into electrical energy or for the control of electrical energy by such radiation
- H01L27/144—Devices controlled by radiation
- H01L27/146—Imager structures
- H01L27/14643—Photodiode arrays; MOS imagers
- H01L27/14645—Colour imagers
Definitions
- the present invention relates to demosacing for use on an image captured by an image sensor.
- Image sensors also known as imagers, were developed in the late 1960s and early 1970s primarily for television image acquisition, transmission, and display.
- An imager absorbs incident radiation of a particular wavelength (such as optical photons, x-rays, or the like) and generates an electrical signal corresponding to the absorbed radiation.
- semiconductor- based imagers including charge coupled devices (CCDs), photodiode arrays, charge injection devices (CIDs), hybrid focal plan arrays, and CMOS imagers.
- These imagers typically consist of an array of pixels containing photosensors, where each pixel produces a signal corresponding to the intensity of light impinging on that element when an image is focused on the array. These signals may then be stored, for example, to display a corresponding image on a monitor or otherwise be used to provide information about the optical image.
- the photosensors are typically phototransistors, photoconductors or photodiodes. The magnitude of the signal produced by each pixel, therefore, is proportional to the amount of light impinging on the photosensor.
- each pixel must be sensitive only to one color or spectral band.
- a color filter array CFA is typically placed in front of the pixels so that each pixel measures the light of the color of its associated filter.
- each pixel of a color image sensor is covered with either a red, green or blue filter, according to a specific pattern (e.g., Bayer pattern).
- CMOS or CCD image sensors the color filters are • integrated with the sensors.
- a common example of a color filter pattern is the tiled color filter array illustrated in U.S. Patent No. 3,971,065, (the disclosure of which is incorporated by reference herein) and commonly referred to as "the Bayer pattern" color filter.
- the Bayer pattern 100 is an array of repeating red (R), green (G), and blue (B) filters causing pixels underneath the filters to be red, blue, and green pixels, respectively.
- red, green and blue pixels are arranged so that alternating red and green pixels are on a first row 105 of an array, and alternating blue and green pixels are on a next row 110. These alternating rows are repeated throughout the array.
- the image sensor is read out, line by line, the pixel sequence for the first line reads GRGRGR etc., and then the alternate line sequence reads BGBGBG etc. This output is called sequential RGB or sRGB.
- the Bayer patterned filters may be formed over an array 200 of pixel sensors 205. Specifically, an array 200 of pixel sensors 205 is formed on a semiconductor substrate 210.
- Each pixel sensor 205 has a photosensitive element 215, which may be any photon-to-charge converting device, such as a photogate, photoconductor or photodiode.
- the color filter array 220 may be formed over an upper level metal layer 225 in the array 200, separated from the photosensors 215 by various metallization and insulating layers such as an interlevel dielectric layer (ILD) 235 containing insulator metallization patterns and a passivation layer 230.
- the metal layer 225 may be opaque and used to shield the areas of the pixels that are not light sensitive.
- Convex lenses 240 are formed over the color filters 220. In operation, incident light is focused by the lenses 240 through the filters 220 to the photosensitive element 215.
- pixel sensor cell 115 is associated with a green filter, which causes pixel sensor cell 115 to sense green light and produce a signal which represents only green light.
- a value may be interpolated from the neighboring red pixel sensor cells 120 and 125 and the neighboring blue pixel sensor cells 130 and 135, respectively. If demosaicing is not performed correctly, the resulting image may suffer from the inclusion of highly visible color artifacts.
- the value used for a first color is based on a sensed color (i.e., the light sensed by the pixel's sensor 205) and the values for the remaining two colors are based on an interpolated value from the sensed values of corresponding neighboring pixels.
- a sensed color i.e., the light sensed by the pixel's sensor 205
- the values for the remaining two colors are based on an interpolated value from the sensed values of corresponding neighboring pixels.
- Each of the sensed values are representative of the color value at the center of the pixel.
- Each of the interpolated color values are also representative of the value at the center of the pixel.
- the interpolated signal is inherently of a lower quality than the originally sensed signal. For example, an interpolated red color value at the center of a pixel would be different than a sensed red color value for the center of the same pixel.
- Demosaicing methods which typically only reconstruct the missing color components, may result in artifacts such as so-called zipper effects and random color dots. Additionally, there exists a trade off between filtering for noise and the blurring of sharp edges. Interpolating a pixel using the surrounding pixel values tends to blur the image when pixels at an edge are interpolated. Furthermore, this also leads to a decrease in the sharpness of the image at the edges. This is mainly to due the averaging of pixels which are across an edge contained within the image. The demosaicing method averages pixels so that a pixel at an edge does not have the sharpness desired of an edge pixel. Accordingly, there is a desire and need for color plane interpolation that provides improved demosaicing techniques that detect and account for image edges when calculating interpolation values. BRIEF SUMMARY OF THE INVENTION
- a method and apparatus for color plane interpolation are provided which interpolates the color values of pixels differently depending on an edge direction and whether a pixel is at an edge within an image.
- the use of the edge detection during the interpolation of each of the colors present in the color pattern helps reduce some of the disadvantages of the loss of image sharpness abundant in known demosaicing techniques.
- FIG.1 is a top-down illustration of a conventional color filter array
- FIG.2 is a cross-sectional view of a portion of a pixel array having a color filter array
- FIGs. 3A-B are exemplary top views of a five by five pixel kernel used in an exemplary interpolation scheme of the present invention.
- FIG.4 is a block diagram of a three by three kernel that is used in accordance with an exemplary embodiment of the invention.
- FIG.5 is a block diagram of a CMOS imager integrated circuit (IC) having a pixel array constructed according to an exemplary embodiment of the invention
- FIG.6 is an illustration of a processor system having a CMOS imager according to the present invention.
- FIG. 7 is a flow chart illustrating the process of one embodiment of the present invention.
- pixel refers to a picture element unit cell containing a photosensor and transistors for converting light radiation to an electrical signal.
- a representative pixel is illustrated in the figures and description herein and, typically, fabrication of all pixels in an imager will proceed simultaneously in a similar fashion.
- the following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined by the appended claims.
- FIGs. 3A, 3B are top-down illustrations of a partial pixel array 300 of an imager in accordance with embodiments of the invention.
- FIGs.3 A, 3B illustrates a five by five kernel (discussed below in more detail) used in the present invention.
- each pixel e.g., pixel 23, senses a single color. Accordingly, each green (G) pixel provides a sensed value for green. Similarly, each blue (B) pixel provides a sensed value for blue and each red (R) pixel provides a sensed value for red.
- the values for each of the remaining colors is interpolated to reconstruct the color image. Additionally, as discussed below, the sensed pixel value may also be replaced by an interpolated value. Interpolation of the red, green and blue values eliminate the imbalance resulting from the use of sensed values for one of the colors and interpolated values of the remaining two colors (prevalent in prior art systems).
- the five by five kernel of surrounding pixels is used in the following interpolation calculations, though in many of the interpolation calculations only pixels in the three by three kernel are needed.
- other kernel sizes may also be employed in the invention.
- a three by three interpolated kernel within the five by five kernel, is first defined and color values for pixel within the kernel are then calculated. Next, an edge direction is determined. Depending on the pixel color and the edge direction, some values of the three by three kernel may be recalculated. The three by three kernel is then used determine whether an edge is present for the selected pixel. Then, based on whether an edge is present and the pixel color final pixel, interpolation values are calculated. Different interpolation parameters are applied in the interpolation process based on whether a processed pixel is at an image edge, as well as the sensed color of the pixel. [0028] The interpolation process of the invention can be more clearly understood from the FIG.
- step 701 a pixel is selected for processing. Throughout the description representative pixel 45 illustrated in FIGs. 3A, 3B and as the center pixel in FIG. 4, is used as the exemplary pixel being processed.
- step 703 the three by three kernel surrounding the selected pixel is defined (e.g., pixels 34, 35, 36, 44, 45, 46, 54, 55, and 56 in FIGs. 3A and 3B).
- step 705 it is determined whether pixel 45 is a sensed green pixel. If pixel 45 is green, in step 707 the green values for the three by three kernel are calculated in accordance with equation set (1) below using actual sensed pixel values from the pixels within the three by three kernel.
- the three by three kernel is calculated using the actual values of green pixels 25, 34, 36, 45, 47, 54, 56 and 65 illustrated in FIG.3A.
- the three by three kernel is calculated using the following equations:
- the same kernel values are interpolated using median interpolation of actual sensed values of surrounding green pixels.
- the three by three interpolated kernel is green, in this exemplary embodiment, because as previously noted in a Bayer CFA, the green pixels are most prevalent and influential in the image.
- the values of pixels in the three by three. kernel gll, gl2, gl3, g21, g22, g23, g31, g32, g33, illustrated in FIG. 4, are a calculation of the green values of the pixels in the three by three kernel illustrated in FIGs. 3A and 3B.
- the green kernel will be used during edge detection and in the calculation of the correction terms.
- One of ordinary skill in the art would appreciate that other pixel colors may be used to calculate the three by three interpolated kernel or that the kernel may be of a different size.
- step 711 the calculated three by three green kernel is used to determine an edge direction.
- the values of four edge directions are calculated in step 711.
- Dashed lines 2, 5, 8 and 11, illustrated in FIG. 4, are the four directions for which edge direction calculations are done.
- the remaining dashed lines are only exemplary for use in calculating the four edge direction values.
- the sum of each dashed line is calculated in equation set (3) below:
- the edge direction value DIRx is calculated for each of the direction lines 2, 5, 8, and 11.
- the following set of equations (4) are used to calculate the four edge direction values: .
- the direction having the greatest DIR value is determined to be the edge direction and is determined in step 711.
- step 715 the edge direction was not vertical, the value of g22 remains unchanged in step 719. If in step 713 the pixel was determined to be green, the value of g22 remains unchanged.
- step 723 if processed pixel 45 is green, the difference values are calculated in step 725 according to equation set (5) below:
- step 723 If it is determined in step 723 that the processed pixel 45 is not green, the difference values are calculated in step 727 according to equation set (6) below:
- step 729 it is determined whether the maximum difference value among the calculated difference values is greater than a predetermined edge threshold value using an equation (7):
- step 731 If the maximum difference value is less than the threshold, no edge is identified as present (step 731), however, if the maximum difference value is greater than the threshold, an edge is identified as present (step 733).
- the edge threshold may be set to be application specific or set based on a user's preferences. [0033] If no edge is detected in step 729, it is again determined if processed pixel 45 is green. If processed pixel 45 is green, in step 739 a eighth set (8) of equations is used to calculated the interpolated red, green and blue values of pixel 45. The eighth set (8) of interpolation equations is:
- R45 (P3S + PS5 + P 37 + P5?)/4;
- G45 (P45+p47+Ps4+P 3 6)/4, where R45, ds, and B45 are the red, green and blue interpolated values for processed pixel 45. Note that even though pixel 45 is a green pixel, the value is interpolated using the actual sensed value of P45 as well as the actual values of other green pixel around it.
- step 735 If in step 735 it is determined that processed pixel 45 is not green, in step 741 it is determined if processed pixel 45 is blue. If processed 45 is blue, a ninth (9) set of equations is used to calculate the red, green and blue values of pixel 45 in step 747.
- the ninth set of equations interpolation is:
- R45, G45, and B45 are the red, green and blue interpolated values for pixel 45. If in step 741, processed pixel 45 is not blue, pixel 45 is red and a tenth (10) set of equations is used to calculate the red, green and blue values of pixel 45 in step 749.
- the tenth set of interpolation equations is:
- R 45 , G45, and B4s are the red, green and blue interpolated values for pixel 45.
- step 737 If an edge was detected in step 729 and processed pixel 45 is green in step 737, the interpolation values of the red, green and blue values of pixel 45 are calculated using an eleventh (11) set of equations.
- the eleventh set of interpolation equations is:
- R45, G4s, and B45 are the red, green and blue interpolated values for pixel 45 and DELTAbiue and DELTAred are correction terms.
- step 745 it is determined if the processed pixel 45 is blue. If processed pixel 45 is blue, a twelfth (12) set of equations is equations is used in step 751 to calculate the red, green and blue interpolation values.
- the twelfth set of interpolation equations is:
- the thirteenth set of interpolation equations is:
- the DELTA correction values for the red and blue pixels are based on the edge direction and the green three by three interpolated kernel.
- the invention provides a new and novel system and method for color plane interpolation, in which interpolated values for green, red and blue are calculated based on not only the sensed values obtained but also whether a processed pixel is at an image image.
- the imbalance between sensed values and interpolated values is removed by interpolating the values for each of the colors.
- edge detection and correction terms are utilized to minimize blurring of image edges.
- FIG. 5 illustrates a block diagram of a CMOS imager integrated circuit (IC) 500 having a pixel array 505 containing a plurality of pixels arranged in rows and columns, including a region 510 with, for example, two green pixels (G), one blue pixel (B), and one red pixel (R) arranged in a Bayer pattern.
- the pixels of each ' row in array 505 are all turned on at the same time by row select lines 515, and the pixels of each column are selectively output by respective column select lines 520.
- the row lines 515 are selectively activated by a row driver 525 in response to row address decoder 530.
- the column select lines 520 are selectively activated by a column selector 535 in response to column address decoder 540.
- the pixel array 505 is operated by the timing and control circuit 545, which controls address decoders 530, 540 for selecting the appropriate row and column lines for pixel signal readout.
- the pixel column signals which typically include a pixel reset signal (Vrst) and a pixel image signal (Vsig), are read by a sample and hold circuit 550 associated with the column selector 535.
- a differential signal (Vrst — Vsig) is produced by differential amplifier 555 for each pixel that is amplified and digitized by analog-to- digital converter 570 (ADC).
- ADC analog-to- digital converter 570
- the analog-to-digital converter 570 supplies the digitized pixel signals to an image processor 575, which performs the processing method 700 of the invention.
- image processor 575 can output an image file in a raw format as received from the analog-to-digital converter 570 which can be received and processed in accordance with the invention by a different processor.
- a typical processor system which includes an imager device 605 having a pixel array in which the pixels signals are interpolated according to the present invention is illustrated generally at 600 in FIG. 6.
- the imager device 605 produces an output image from signals supplied from the pixel array.
- a processor system is exemplary of a system receiving the output of a CMOS imager device. Without being limiting, such a system could include a computer system, camera system, scanner, machine vision system, medical sensor system (such as medical pill sensors), and automotive diagnostic system, and other imaging systems, all of which can utilize the present invention.
- the image processor 575 in the imager device 605 provides a progressed image signal, the invention can performed in. the processing system 600.
- a processor system 600 such as a computer system, for example generally comprises a central processing unit (CPU) 610, for example, a microprocessor, that communicates with an input/output (I/O) device 615 over a bus 620.
- the imager device 605 also communicates with the system over bus 620 or other communication link.
- the processor-based system 600 also includes random access memory (RAM) 625, and can include removable memory 630, such as flash memory, which also communicates with the CPU 610 over the bus 620.
- the imaging device 600 may be combined with a processor, such as a CPU, digital signal processor, or microprocessor, with or without memory storage on a single integrated circuit or on a different chip than the processor.
- a program embodying the processing method may be stored on a carrier medium which may include RAM, floppy disk, data transmission, compact disk, etc. and then be executed by an associated processor which receives and processes an image file in accordance with the invention or the processing method may be embodied in a circuit or a combination thereof.
- the invention may be implemented as a plug-in for existing software applications or it may used on its own.
- the invention is not limited to the carrier mediums specified herein and the invention may be implemented using any carrier medium as known in the art.
Abstract
A method and apparatus for color plane interpolation are provided which interpolates the color values of pixels differently depending on an edge direction and whether a pixel is at an edge within an image. The use of the edge detection during the interpolation of each of the colors present in the color pattern helps reduce some of the disadvantages of the loss of image sharpness abundant in known demosaicing techniques.
Description
METHOD AND APPARATUS PROVIDING COLOR INTERPOLATION
IN COLOR FILTER ARRAYS USING EDGE DETECTION
AND CORRECTION TERMS
[0001] The present invention relates to demosacing for use on an image captured by an image sensor.
BACKGROUND OF THE INVENTION
[0002] Image sensors, also known as imagers, were developed in the late 1960s and early 1970s primarily for television image acquisition, transmission, and display. An imager absorbs incident radiation of a particular wavelength (such as optical photons, x-rays, or the like) and generates an electrical signal corresponding to the absorbed radiation. There are a number of different types of semiconductor- based imagers, including charge coupled devices (CCDs), photodiode arrays, charge injection devices (CIDs), hybrid focal plan arrays, and CMOS imagers.
[0003] These imagers typically consist of an array of pixels containing photosensors, where each pixel produces a signal corresponding to the intensity of light impinging on that element when an image is focused on the array. These signals may then be stored, for example, to display a corresponding image on a monitor or otherwise be used to provide information about the optical image. The photosensors are typically phototransistors, photoconductors or photodiodes. The magnitude of the signal produced by each pixel, therefore, is proportional to the amount of light impinging on the photosensor.
[0004] To allow the photosensors to capture a color image, the photosensors must be able to separately detect, for example when using a Bayer pattern, red (R) photons, green (G) photons and blue (B) photons. Accordingly, each pixel must be sensitive only to one color or spectral band. For this, a color filter array (CFA) is
typically placed in front of the pixels so that each pixel measures the light of the color of its associated filter. Thus, each pixel of a color image sensor is covered with either a red, green or blue filter, according to a specific pattern (e.g., Bayer pattern).
[0005] For most low cost CMOS or CCD image sensors, the color filters are • integrated with the sensors. A common example of a color filter pattern is the tiled color filter array illustrated in U.S. Patent No. 3,971,065, (the disclosure of which is incorporated by reference herein) and commonly referred to as "the Bayer pattern" color filter.
[0006] As shown in FIG. 1, the Bayer pattern 100 is an array of repeating red (R), green (G), and blue (B) filters causing pixels underneath the filters to be red, blue, and green pixels, respectively. In the Bayer pattern 100, red, green and blue pixels are arranged so that alternating red and green pixels are on a first row 105 of an array, and alternating blue and green pixels are on a next row 110. These alternating rows are repeated throughout the array. Thus, when the image sensor is read out, line by line, the pixel sequence for the first line reads GRGRGR etc., and then the alternate line sequence reads BGBGBG etc. This output is called sequential RGB or sRGB.
[0007] In the Bayer pattern 100, sampling rates for all three basic colors are adjusted according to the acuity of the human visual system. That is, green color, to which the human eye is most sensitive and responsive, is sensed with a larger number of sensors, whereas blue and red color, for which the human vision has less resolution, are sensed with a fewer number of sensors. This is why in the Bayer pattern, the green-sensitive elements occur at every other array position, while the red-sensitive elements and the blue-sensitive elements occur at every fourth array position.
[0008] As shown in FIG. 2 in a solid state image sensor, the Bayer patterned filters may be formed over an array 200 of pixel sensors 205. Specifically, an array 200 of pixel sensors 205 is formed on a semiconductor substrate 210. Each pixel sensor 205 has a photosensitive element 215, which may be any photon-to-charge converting device, such as a photogate, photoconductor or photodiode. The color filter array 220 may be formed over an upper level metal layer 225 in the array 200, separated from the photosensors 215 by various metallization and insulating layers such as an interlevel dielectric layer (ILD) 235 containing insulator metallization patterns and a passivation layer 230. The metal layer 225 may be opaque and used to shield the areas of the pixels that are not light sensitive. Convex lenses 240 are formed over the color filters 220. In operation, incident light is focused by the lenses 240 through the filters 220 to the photosensitive element 215.
[0009] For a Bayer pattern filter, values for red, green and blue are necessary for each pixel. Since each pixel sensor cell is only sensing one color, values for the remaining two colors are calculated by interpolation from the neighboring pixels that are sensing the missing colors. This color plane interpolation is known as demosaicing. For example, with reference to FIG. 1, pixel sensor cell 115 is associated with a green filter, which causes pixel sensor cell 115 to sense green light and produce a signal which represents only green light. In order to obtain an approximation of the amount of red and blue light for pixel sensor cell 115, a value may be interpolated from the neighboring red pixel sensor cells 120 and 125 and the neighboring blue pixel sensor cells 130 and 135, respectively. If demosaicing is not performed correctly, the resulting image may suffer from the inclusion of highly visible color artifacts.
[0010] The article entitled "Color Plane Interpolation Using Alternating Projections" published in the IEEE Transactions on Image Processing, Vol. II, No.9
in September 2002 and written by Bahadir K. Gunturk, Yucel Altunbasak and Russell M. Mersereau (the disclosure of which is incorporated by reference herein) compares several demosaicing techniques. As described each of these demosaicing techniques have their respective advantages and disadvantages.
[0011] As described, for each pixel, the value used for a first color is based on a sensed color (i.e., the light sensed by the pixel's sensor 205) and the values for the remaining two colors are based on an interpolated value from the sensed values of corresponding neighboring pixels. Each of the sensed values are representative of the color value at the center of the pixel. Each of the interpolated color values are also representative of the value at the center of the pixel. The interpolated signal is inherently of a lower quality than the originally sensed signal. For example, an interpolated red color value at the center of a pixel would be different than a sensed red color value for the center of the same pixel.
[0012] Demosaicing methods, which typically only reconstruct the missing color components, may result in artifacts such as so-called zipper effects and random color dots. Additionally, there exists a trade off between filtering for noise and the blurring of sharp edges. Interpolating a pixel using the surrounding pixel values tends to blur the image when pixels at an edge are interpolated. Furthermore, this also leads to a decrease in the sharpness of the image at the edges. This is mainly to due the averaging of pixels which are across an edge contained within the image. The demosaicing method averages pixels so that a pixel at an edge does not have the sharpness desired of an edge pixel. Accordingly, there is a desire and need for color plane interpolation that provides improved demosaicing techniques that detect and account for image edges when calculating interpolation values.
BRIEF SUMMARY OF THE INVENTION
[0013] A method and apparatus for color plane interpolation are provided which interpolates the color values of pixels differently depending on an edge direction and whether a pixel is at an edge within an image. The use of the edge detection during the interpolation of each of the colors present in the color pattern helps reduce some of the disadvantages of the loss of image sharpness abundant in known demosaicing techniques.
BRIEF DESCRIPTION OF THE DRAWINGS
[0014] The foregoing and other advantages and features of the invention will become more apparent from the detailed description of exemplary embodiments provided below with reference to the accompanying drawings in which: "*
[0015] FIG.1 is a top-down illustration of a conventional color filter array;
[0016] FIG.2 is a cross-sectional view of a portion of a pixel array having a color filter array;
[0017] FIGs. 3A-B are exemplary top views of a five by five pixel kernel used in an exemplary interpolation scheme of the present invention;
[0018] FIG.4 is a block diagram of a three by three kernel that is used in accordance with an exemplary embodiment of the invention;
[0019] FIG.5 is a block diagram of a CMOS imager integrated circuit (IC) having a pixel array constructed according to an exemplary embodiment of the invention;
[0020] FIG.6 is an illustration of a processor system having a CMOS imager according to the present invention; and
[0021] FIG. 7 is a flow chart illustrating the process of one embodiment of the present invention.
DETAILED DESCRIPTION OF THE INVENTION
[0022] In the following detailed description, reference is made to the accompanying drawings which form a part hereof, and in which is shown by way of illustration specific embodiments in which the invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized/ and that structural, logical and electrical changes may be made without departing from the spirit and scope of the present invention.
[0023] The term "pixel" refers to a picture element unit cell containing a photosensor and transistors for converting light radiation to an electrical signal. For purposes of illustration, a representative pixel is illustrated in the figures and description herein and, typically, fabrication of all pixels in an imager will proceed simultaneously in a similar fashion. The following detailed description is, therefore, not to be taken in a limiting sense, and the scope of the present invention is defined by the appended claims.
[0024] It should be understood that reference to a CMOS imager in accordance with the present invention is made for the sole purpose of describing just one example of the present invention. It should be readily apparent that the present invention is not limited to CMOS imagers, but also applies to CCD and other imagers that employ color filters over pixels. Additionally, the current invention is described using a standard three color Bayer pattern. It should be understood that the present invention is not limited to the standard three color Bayer pattern but may be applied to color spaces which use different colors or which use more, or less, than three colors.
[0025] FIGs. 3A, 3B are top-down illustrations of a partial pixel array 300 of an imager in accordance with embodiments of the invention. Specifically, FIGs.3 A, 3B illustrates a five by five kernel (discussed below in more detail) used in the present invention. As previously noted, each pixel, e.g., pixel 23, senses a single color. Accordingly, each green (G) pixel provides a sensed value for green. Similarly, each blue (B) pixel provides a sensed value for blue and each red (R) pixel provides a sensed value for red. For each pixel of a particular color, the values for each of the remaining colors is interpolated to reconstruct the color image. Additionally, as discussed below, the sensed pixel value may also be replaced by an interpolated value. Interpolation of the red, green and blue values eliminate the imbalance resulting from the use of sensed values for one of the colors and interpolated values of the remaining two colors (prevalent in prior art systems).
[0026] As shown in FIGs. 3A, 3B for a particular pixel 45 being processed, the five by five kernel of surrounding pixels is used in the following interpolation calculations, though in many of the interpolation calculations only pixels in the three by three kernel are needed. However, other kernel sizes may also be employed in the invention.
[0027] Generally, for each selected pixel, a three by three interpolated kernel, within the five by five kernel, is first defined and color values for pixel within the kernel are then calculated. Next, an edge direction is determined. Depending on the pixel color and the edge direction, some values of the three by three kernel may be recalculated. The three by three kernel is then used determine whether an edge is present for the selected pixel. Then, based on whether an edge is present and the pixel color final pixel, interpolation values are calculated. Different interpolation parameters are applied in the interpolation process based on whether a processed pixel is at an image edge, as well as the sensed color of the pixel.
[0028] The interpolation process of the invention can be more clearly understood from the FIG. 7 flow chart which illustrates the overall interpolation process. In step 701, a pixel is selected for processing. Throughout the description representative pixel 45 illustrated in FIGs. 3A, 3B and as the center pixel in FIG. 4, is used as the exemplary pixel being processed. Next in step 703, the three by three kernel surrounding the selected pixel is defined (e.g., pixels 34, 35, 36, 44, 45, 46, 54, 55, and 56 in FIGs. 3A and 3B). In step 705 it is determined whether pixel 45 is a sensed green pixel. If pixel 45 is green, in step 707 the green values for the three by three kernel are calculated in accordance with equation set (1) below using actual sensed pixel values from the pixels within the three by three kernel. FIG. 4 illustrates the locations within the three by three kernel for the calculated green values gll, gl2, gl3, g21, g22, g23, g31, g32, g33 for pixels within the three by three kernel. The three by three kernel is calculated using the actual values of green pixels 25, 34, 36, 45, 47, 54, 56 and 65 illustrated in FIG.3A. The three by three kernel is calculated using the following equations:
where "P" is the pixel value and "xy" is the pixel number shown in FIGs. 3A-B. As shown in FIGs.3A-B, "x" delineates the row and "y" delineates the column. If pixel 45 is either red or blue as shown in FIG. 3B, in step 709, the three by three kernel is calculated using the values of green pixels 24, 26, 33, 35, 37, 44, 46,.53, 55, 57, 64 and 65 in accordance with equation set (2) below. The three by three green kernel is then calculated using the following equations:
(2)
In this exemplary embodiment, the same kernel values are interpolated using median interpolation of actual sensed values of surrounding green pixels. The three by three interpolated kernel is green, in this exemplary embodiment, because as previously noted in a Bayer CFA, the green pixels are most prevalent and influential in the image. The values of pixels in the three by three. kernel gll, gl2, gl3, g21, g22, g23, g31, g32, g33, illustrated in FIG. 4, are a calculation of the green values of the pixels in the three by three kernel illustrated in FIGs. 3A and 3B. The green kernel will be used during edge detection and in the calculation of the correction terms. One of ordinary skill in the art would appreciate that other pixel colors may
be used to calculate the three by three interpolated kernel or that the kernel may be of a different size.
[0029] Next in step 711, illustrated in FIG. 7, the calculated three by three green kernel is used to determine an edge direction. The values of four edge directions are calculated in step 711. Dashed lines 2, 5, 8 and 11, illustrated in FIG. 4, are the four directions for which edge direction calculations are done. The remaining dashed lines are only exemplary for use in calculating the four edge direction values. The sum of each dashed line is calculated in equation set (3) below:
[0030] The edge direction value DIRx, where "DIR" is the value and "x" is the direction line number, is calculated for each of the direction lines 2, 5, 8, and 11. The following set of equations (4) are used to calculate the four edge direction values: .
DIR2
DIR5
DIR8 (4)
The direction having the greatest DIR value is determined to be the edge direction and is determined in step 711.
[0031] Next in step 713 illustrated in FIG. 7, it is determined if pixel 45 undergoing processing is green. If the pixel 45 is not green, in step 713, it is determined whether the edge direction was in the direction of line 2 (horizontal) in step 714. If the edge direction was horizontal, g22 is recalculated to account for the edge direction, illustrated in FIG. 3B. The value of g22 is recalculated using the equation g22=(g21+g23)/2. If in step 714, the edge direction is not horizontal, it is next determined if the edge direction -was in the direction of line 5 (vertical) in step 715. If the direction was determined to be in vertical direction the pixel value of g22 is recalculated using the equation g22=(gl2+g32)/2. However, if in step 715 the edge direction was not vertical, the value of g22 remains unchanged in step 719. If in step 713 the pixel was determined to be green, the value of g22 remains unchanged.
[0032] Next, in order to determine if an edge is present for processed pixel 45, difference values are calculated between pairs of the same color pixels for all of the pixels that may be used in calculating the interpolated values. In step 723 if processed pixel 45 is green, the difference values are calculated in step 725 according to equation set (5) below:
If it is determined in step 723 that the processed pixel 45 is not green, the difference values are calculated in step 727 according to equation set (6) below:
In step 729, it is determined whether the maximum difference value among the calculated difference values is greater than a predetermined edge threshold value using an equation (7):
MAX (Di ... D7) > Edge Threshold. (7)
If the maximum difference value is less than the threshold, no edge is identified as present (step 731), however, if the maximum difference value is greater than the threshold, an edge is identified as present (step 733). The edge threshold may be set to be application specific or set based on a user's preferences.
[0033] If no edge is detected in step 729, it is again determined if processed pixel 45 is green. If processed pixel 45 is green, in step 739 a eighth set (8) of equations is used to calculated the interpolated red, green and blue values of pixel 45. The eighth set (8) of interpolation equations is:
R45=(P3S + PS5 + P37 + P5?)/4;
+ P46 +Pβ6 +P<54)/4; and (8)
G45=(P45+p47+Ps4+P36)/4,
where R45, ds, and B45 are the red, green and blue interpolated values for processed pixel 45. Note that even though pixel 45 is a green pixel, the value is interpolated using the actual sensed value of P45 as well as the actual values of other green pixel around it.
[0034] If in step 735 it is determined that processed pixel 45 is not green, in step 741 it is determined if processed pixel 45 is blue. If processed 45 is blue, a ninth (9) set of equations is used to calculate the red, green and blue values of pixel 45 in step 747. The ninth set of equations interpolation is:
(9)
where R45, G45, and B45 are the red, green and blue interpolated values for pixel 45. If in step 741, processed pixel 45 is not blue, pixel 45 is red and a tenth (10) set of equations is used to calculate the red, green and blue values of pixel 45 in step 749.
The tenth set of interpolation equations is:
(10)
[0035] If an edge was detected in step 729 and processed pixel 45 is green in step 737, the interpolation values of the red, green and blue values of pixel 45 are calculated using an eleventh (11) set of equations. The eleventh set of interpolation equations is:
(11)
where R45, G4s, and B45 are the red, green and blue interpolated values for pixel 45 and DELTAbiue and DELTAred are correction terms. The values of DELTAbiue and DELTAred are calculated based on which direction the edge was determined to be in step 711. If the edge direction is in the horizontal line 2 direction, DELTAbiue = g22- (g21+g23)/2 and DELTAred = 0 and if the edge direction is determine to be vertical line 5 direction, DELTAbiue = 0 and DELTAred = g22-(gl2+g32)/2. Lastly, if the edge direction is determined to be in a diagonal direction 11 or 8, DELTAbiue = g22- (g21+g23)/2 and DELTAred = g22-(gl2+g32)/2.
[0036] If in step 737 it is determined that processed pixel 45 is not green, in step 745 it is determined if the processed pixel 45 is blue. If processed pixel 45 is blue, a twelfth (12) set of equations is equations is used in step 751 to calculate the red, green and blue interpolation values. The twelfth set of interpolation equations is:
■where R45, G45, and B45 are the red, green and blue interpolated values for pixel 45 and the value of DELTAredis calculated using the equation DELTAred = g22- (gll+gl3+g31+g33)/4. If in step 745, processed pixel 45 is not blue, a thirteenth (13) set of equations is used to calculate the red, green and blue values for red processed pixel 45 in step 753. The thirteenth set of interpolation equations is:
where R45, G45, and B45 are the red, green and blue interpolated values for pixel 45 and the value of DELTAbiueis calculated using the equation DELTAbiue = g22- (gll+gl3+g31+g33)/4). The DELTA correction values for the red and blue pixels are based on the edge direction and the green three by three interpolated kernel.
[0037] As described, the invention provides a new and novel system and method for color plane interpolation, in which interpolated values for green, red and blue are calculated based on not only the sensed values obtained but also whether a processed pixel is at an image image. According to the invention, the imbalance between sensed values and interpolated values is removed by interpolating the values for each of the colors. Further, edge detection and correction terms are utilized to minimize blurring of image edges.
[0038] FIG. 5 illustrates a block diagram of a CMOS imager integrated circuit (IC) 500 having a pixel array 505 containing a plurality of pixels arranged in rows and columns, including a region 510 with, for example, two green pixels (G), one
blue pixel (B), and one red pixel (R) arranged in a Bayer pattern. The pixels of each ' row in array 505 are all turned on at the same time by row select lines 515, and the pixels of each column are selectively output by respective column select lines 520.
[0039] The row lines 515 are selectively activated by a row driver 525 in response to row address decoder 530. The column select lines 520 are selectively activated by a column selector 535 in response to column address decoder 540. The pixel array 505 is operated by the timing and control circuit 545, which controls address decoders 530, 540 for selecting the appropriate row and column lines for pixel signal readout.
[0040] The pixel column signals, which typically include a pixel reset signal (Vrst) and a pixel image signal (Vsig), are read by a sample and hold circuit 550 associated with the column selector 535. A differential signal (Vrst — Vsig) is produced by differential amplifier 555 for each pixel that is amplified and digitized by analog-to- digital converter 570 (ADC). The analog-to-digital converter 570 supplies the digitized pixel signals to an image processor 575, which performs the processing method 700 of the invention. Alternatively, image processor 575 can output an image file in a raw format as received from the analog-to-digital converter 570 which can be received and processed in accordance with the invention by a different processor.
[0041] A typical processor system which includes an imager device 605 having a pixel array in which the pixels signals are interpolated according to the present invention is illustrated generally at 600 in FIG. 6. The imager device 605 produces an output image from signals supplied from the pixel array. A processor system is exemplary of a system receiving the output of a CMOS imager device. Without being limiting, such a system could include a computer system, camera system, scanner, machine vision system, medical sensor system (such as medical pill
sensors), and automotive diagnostic system, and other imaging systems, all of which can utilize the present invention. Alternatively, if the image processor 575 in the imager device 605 provides a progressed image signal, the invention can performed in. the processing system 600.
[0042] A processor system 600, such as a computer system, for example generally comprises a central processing unit (CPU) 610, for example, a microprocessor, that communicates with an input/output (I/O) device 615 over a bus 620. The imager device 605 also communicates with the system over bus 620 or other communication link. The processor-based system 600 also includes random access memory (RAM) 625, and can include removable memory 630, such as flash memory, which also communicates with the CPU 610 over the bus 620. The imaging device 600 may be combined with a processor, such as a CPU, digital signal processor, or microprocessor, with or without memory storage on a single integrated circuit or on a different chip than the processor.
[0043] The invention is not restricted to the above described embodiments. For example, a program embodying the processing method may be stored on a carrier medium which may include RAM, floppy disk, data transmission, compact disk, etc. and then be executed by an associated processor which receives and processes an image file in accordance with the invention or the processing method may be embodied in a circuit or a combination thereof. Additionally, the invention may be implemented as a plug-in for existing software applications or it may used on its own. The invention is not limited to the carrier mediums specified herein and the invention may be implemented using any carrier medium as known in the art.
[0044] It should be understood that other methods of interpolating the values for the green, red and blue colors of the pixels are within the scope of the invention, and the invention is in not limited to the equations or use of the correction terms
described above. Specifically, for example, other surrounding pixels may be used in calculating the interpolated values instead or in addition to those discussed herein. As such, the above description and drawings are only to be considered illustrative of an exemplary embodiment which achieves the features and advantages of the invention. Any modification of the present invention which comes within the spirit and scope of the following claims should be considered part of the invention. Accordingly, the invention is not to be considered as being limited by the foregoing description and drawings, but is only limited by the scope of the appended claims.
Claims
1. A method of color plane interpolation for image pixels, said method comprising the steps of:
selecting a pixel for processing;
forming a kernel of neighboring pixels associated with said selected pixel;
determining whether said selected pixel is at an edge within an image and the direction of such edge; and
generating a value for said selected pixel for each of color of a color pattern based on said neighboring pixels, whether said selected pixel is at said edge and the direction of said edge.
2. The method of claim 1, further comprising the act of calculating pixel values of one color of said color pattern for all of the pixels in said kernel and determining said edge direction using said calculated kernel pixel values of one color.
3. The method of claim 2, further comprising recalculating at least one pixel value in said kernel based on said determined edge direction.
4. The method of claim 1, wherein said color pattern is a Bayer pattern.
5. The method of claim 1, wherein said one color is green.
6. The method of claim 2, wherein said act of recalculating said at least one kernel value is only recalculated if the edge direction is determined to be horizontal or vertical.
7. The method of claim 1, wherein said step of determining whether said edge is present comprises:
calculating difference values between different pairs of pixels of a same color; and
comparing the largest difference value to a predetermined threshold value.
8. The method of claim 7, wherein said pairs of pixels may include said pixel.
9. A color imaging device including a sensing array of image pixel sensors, said device comprising:
a first plurality of image pixel sensors that are used to produce sensed values of a first color;
a second plurality of image pixel sensors that are used to produce sensed values of a second color;
a third plurality of image pixel sensors that are used to produce sensed values of a third color;
interpolation means for using said sensed values of the first color to determine a first set of pixel values for a plurality of image pixels in a pixel kernel, including a selected pixel and neighboring pixels, for said first color, determining whether an edge is present for said selected pixel using said kernel values and calculating interpolation values for said first, second and third colors of said selected pixel based on said kernel values, neighboring pixels and an edge direction; and
creating a representative image using interpolated values of said first, second and third colors.
10. The color imaging device of claim 9, wherein said interpolation means further whether said selected pixel is at said edge.
11. The color imaging device of claim 10, wherein at least one value of said pixel kernel is recalculated based on said edge direction before said interpolation values are calculated.
12. The color imaging device of claim 9, wherein said pixel kernel is a three by three kernel of pixels.
13. The color imaging device of claim 10, wherein said edge directions are calculated as follow:
DIRhorizontai = m«( I S1-S21 , I S2-S31 );
DIRv.rtic-1 = max( I S4-S5 i , I S5-S61 ); DIRdiagonaii = max( I S7-S81 , I S8-S91 ); and
DIRdiagonaπ = max( I SlO-SIl I , I S11-S121 ), where Sl, S2, S3, S4, S5, S6, S7, S8, S9, SlO, SlI and S12 are calculated as follows:
Sl= gll+gl2+gl3 S2=g21+g22+g23
S3=g31+g32+g33;
S4=gll+g21+g31; •
S5=gl2+g22+g32;
S6=gl3+g23+g33;
S7=g21+gll+gl2);
S8=gl3+g22+g31;
S9=g23+g32+g33;
S10=g21+g31+g32;
Sll=gll+gl3+g33;
S12=gl2+gl3+g23, using said kernel where gll, gl2, and gl3 are the pixel values of the first row of said kernel, g21, g22, and g23 are the pixel values of the second row of said kernel and g31, g 32, and g33 are the pixel values of the third row of the kernel and wherein said edge direction is the direction having the greatest value.
14. The color imaging device of claim 9, wherein said edge is detected by calculating difference values of neighboring pixels and comparing a maximum difference value to an edge threshold.
15. The color imaging device of claim 14, wherein said difference values are calculated as follows when said selected pixel is green:
Dl = I PxH, y+l-Px-1, y+l I ;
Ds = I Px+i, y-Px-i, y+21 ; and
where Px, y is the selected pixel and the "x" relates to the row in which the pixel is relative to the selected pixel Px, y and "y" relates to the column in which the pixel is relative to the selected pixel Px, y. '
16. The color imaging device of claim 14, wherein said difference values are calculated as follows when said selected pixel is red or blue:
D2 = I Px+1, y+1 -Px-I, y-1 I ;
17. The color imaging device of claim 12, wherein said three by three kernel is calculated, when said pixel is green, as follows: gl2-MEDIAN(Pχ-2,y/ Px-i,y-i, Px, y, Px, y«);
18. The color imaging device of claim 12, wherein said three by three kernel is calculated, when said pixel is red or blue, as follows:
gll= MEDIAN(Px-I, y-2, Px-2,y-i, Px-i,y, Pχ,y-i);
19. The color imaging device of claim 9, wherein said calculated interpolation values for said first, second and third colors of said selected pixel are calculated as follows when said selected pixel is green and no edge is present:
Rx, y + Px+1, y + Px-I, y+2 + Px+1, y«)/4;
Bx, i + Px, y+i +Pχ+2, y+i +Px+2, y-i)/4; and
where Px, y is the selected pixel and the "x" relates to the row in which the pixel is relative to the selected pixel Px, y and "y" relates to the column in which the pixel is relative to the selected pixel Px, y.
20. The color imaging device of claim 9, wherein said calculated interpolation values for said first, second and third colors of said selected pixel are calculated as follows when said selected pixel is blue and no edge is present:
Rx, y=(Py+1, x+l + Px+1, y-1 + Px-I, y+1 + Psx-1, y-1 )/4;
Bx, Px, y-1 + Px, y+2 +Px+2, y)/4; and
where Px, y is the selected pixel and the "x" relates to the row in which the pixel is relative to the selected pixel Px, y and "y" relates to the column in which the pixel is relative to the selected pixel Px, y.
21. The color imaging device of claim 9, wherein said calculated interpolation values for said first, second and third colors of said selected pixel are calculated as follows ■when said selected pixel is red and no edge is present:
Bx, y=(Py+l, x+1 + Px+1, y-1 + Px-I, y+1 + P3x-1, y-l)/4;
Rx, Px, y-1 + Px, y+2 +Px+2, y)/4; and
where Px, y is the selected pixel and the "x" relates to the row in which the pixel is relative to the selected pixel Px, y and "y" relates to the column in which the pixel is relative to the selected pixel Px, γ.
22. The color imaging device of claim 9, wherein said calculated interpolation values for said first, second and third colors of said selected pixel are calculated as follows when said selected pixel is green and an edge is present:
Px, y+i)/2+ DELTAbiu.; and
where g22, and correction terms DELTAred and DELTAbiue are calculated based on said kernel values and -where Px, y is the selected pixel and the "x" relates to the row in which the pixel is relative to the selected pixel Px, y and "y" relates to the column in which the pixel is relative to the selected pixel Px, y.
23. The color imaging device of claim 9, wherein said calculated interpolation values for said first, second and third colors of said selected pixel are calculated as follows when said selected pixel is blue and an edge is present:
Px+I, y-1 + Px-I, y+l + Px-I, y-l)/4 + DELTAred,'
24. The color imaging device of claim 9, wherein said calculated interpolation values for said first second and third colors of said selected pixel are calculated as follows when said selected pixel is red and an edge is present:
Px+I, y-l + Px-I, yrt + PX-l,y-l)/4 + DELTAred;
25. A processing system comprising:
an imaging device including an array of image pixel sensors, said device comprising:
a first plurality of image pixel sensors that are used to produce sensed values of a first color;
a second plurality of image pixel sensors that are used to produce sensed values of a second color;
a third plurality of image pixel sensors that are used to produce sensed values of a third color;
interpolation processing circuit for using said sensed values of the first color to determine a first set of pixel values for a plurality of image pixels in a pixel kernel, including a selected pixel and neighboring pixels, for said first color, determining whether an edge is present for said selected pixel using said kernel values and calculating interpolation values for said first, second and third colors of said selected pixel based on said kernel values, neighboring pixels and an edge direction of said selected pixel.
26. The processing system of claim 25, wherein said interpolation means further determines whether said selected pixel is at said edge.
27. The processing system of claim 25, wherein said edge is calculated using said kernel values.
28. The processing system of claim 25, wherein at least one value of said pixel kernel is recalculated based on said edge direction.
29. A method of processing a digital color signal comprising the acts of:
obtaining a sensed value from each of a plurality of pixel sensors, the plurality of pixel sensors corresponding to applied light with wavelengths in one of a red, blue and green spectrum; and
interpolating a red value, a blue value and a green value for each of said plurality of pixel sensors based on edge detection and correction terms determined using said sensed values.
30. The method of claim 29, wherein said interpolation of said red, green and blue values further comprises:
calculating a green kernel of interpolation values using a pixel values of neighboring green pixel sensors for each of said plurality of pixel sensors; and
determining whether each of said pixel sensors is at an edge.
31. The method of claim 29, further comprising determining an edge direction using said green kernel of interpolation values.
32. The method of claim 30, wherein said step of determining whether said pixel is at an edge comprises:
calculating difference values between different pairs of neighboring pixels of a same color; and comparing the largest difference value to a predetermined threshold value.
33. A computer readable media for storing computer instructions for enabling a computer to perform color plane interpolation for a color pattern including the steps of:
obtaining a sensed value from each of a plurality of pixel sensors, the plurality of pixel sensors corresponding to applied light with wavelengths in one of a red, blue and green spectrum; and
interpolating a red value, a blue value and a green value for said plurality of pixel sensors based on edge detection and correction terms determined using said sensed values.
34. The medium of claim 33, wherein said interpolation of said red, green and blue values further comprises:
calculating a green kernel of interpolation values using a set of neighboring green pixel sensors for each of said plurality of pixel sensors; and
determining whether said pixel is at an edge for each of said plurality of pixel sensors.
35. The medium of claim 34, further comprising determining an edge direction using said green kernel of interpolation values.
36. The medium of claim 34, wherein said step of determining whether said pixel is at an edge comprises: calculating difference values between different pairs of neighboring pixels of a same color; and
comparing the largest difference value to a predetermined threshold value.
37. A method of color plane interpolation for image pixels, for a selected pixel said method comprising:
determining whether said selected pixel is at an image edge;
if said selected pixel is at said edge, calculating pixel values for each color of a color pattern based on said neighboring pixels using a first set of parameters; and
if said selected pixel is at not said edge, calculating pixel values for each color of said color pattern based on said neighboring pixels using a second set of parameters.
38. The method of claim 37, wherein said first and second set of parameters are based on the color of said selected pixel.
39. The method of claim 37, wherein said color pattern is a Bayer color pattern.
40. The method of claim 39, wherein said first set of parameters is comprised of a third set of parameters for green pixels, a fourth set of parameters for blue pixels and a fifth set of parameters for red pixels.
41. The method of claim 39, wherein said second set of parameters is comprised of a sixth set of parameters for green pixels, a seventh set of parameters for blue pixels and an eighth set of parameters for red pixels.
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP06845702A EP1969558B1 (en) | 2005-12-29 | 2006-12-19 | Method and apparatus providing color interpolation in color filter arrays using edge detection and correction terms |
AT06845702T ATE520100T1 (en) | 2005-12-29 | 2006-12-19 | METHOD AND APPARATUS FOR PROVIDING COLOR INTERPOLATION IN COLOR FILTER ARRAYS USING EDGE DETECTION AND CORRECTION TERMS |
JP2008548595A JP4893973B2 (en) | 2005-12-29 | 2006-12-19 | Method and apparatus for providing color interpolation in a color filter array using edge detection and correction terms |
CN2006800519591A CN101336438B (en) | 2005-12-29 | 2006-12-19 | Method and apparatus providing color interpolation in color filter arrays using edge detection and correction terms |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US11/319,398 US7830426B2 (en) | 2005-12-29 | 2005-12-29 | Method and apparatus providing color interpolation in color filter arrays using edge detection and correction terms |
US11/319,398 | 2005-12-29 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2007078912A1 true WO2007078912A1 (en) | 2007-07-12 |
Family
ID=37909481
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2006/048200 WO2007078912A1 (en) | 2005-12-29 | 2006-12-19 | Method and apparatus providing color interpolation in color filter arrays using edge detection and correction terms |
Country Status (8)
Country | Link |
---|---|
US (2) | US7830426B2 (en) |
EP (1) | EP1969558B1 (en) |
JP (1) | JP4893973B2 (en) |
KR (1) | KR101007754B1 (en) |
CN (1) | CN101336438B (en) |
AT (1) | ATE520100T1 (en) |
TW (1) | TWI337499B (en) |
WO (1) | WO2007078912A1 (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8330772B2 (en) | 2009-10-20 | 2012-12-11 | Apple Inc. | Image signal processor front-end image data processing system and method |
US8471932B2 (en) | 2010-09-30 | 2013-06-25 | Apple Inc. | Spatial filtering for image signal processing |
US8593483B2 (en) | 2009-10-20 | 2013-11-26 | Apple Inc. | Temporal filtering techniques for image signal processing |
US8817120B2 (en) | 2012-05-31 | 2014-08-26 | Apple Inc. | Systems and methods for collecting fixed pattern noise statistics of image data |
US8872946B2 (en) | 2012-05-31 | 2014-10-28 | Apple Inc. | Systems and methods for raw image processing |
US8917336B2 (en) | 2012-05-31 | 2014-12-23 | Apple Inc. | Image signal processing involving geometric distortion correction |
US8953882B2 (en) | 2012-05-31 | 2015-02-10 | Apple Inc. | Systems and methods for determining noise statistics of image data |
US9014504B2 (en) | 2012-05-31 | 2015-04-21 | Apple Inc. | Systems and methods for highlight recovery in an image signal processor |
US9025867B2 (en) | 2012-05-31 | 2015-05-05 | Apple Inc. | Systems and methods for YCC image processing |
US9031319B2 (en) | 2012-05-31 | 2015-05-12 | Apple Inc. | Systems and methods for luma sharpening |
US9077943B2 (en) | 2012-05-31 | 2015-07-07 | Apple Inc. | Local image statistics collection |
US9105078B2 (en) | 2012-05-31 | 2015-08-11 | Apple Inc. | Systems and methods for local tone mapping |
US9131196B2 (en) | 2012-05-31 | 2015-09-08 | Apple Inc. | Systems and methods for defective pixel correction with neighboring pixels |
US9142012B2 (en) | 2012-05-31 | 2015-09-22 | Apple Inc. | Systems and methods for chroma noise reduction |
US9332239B2 (en) | 2012-05-31 | 2016-05-03 | Apple Inc. | Systems and methods for RGB image processing |
US11089247B2 (en) | 2012-05-31 | 2021-08-10 | Apple Inc. | Systems and method for reducing fixed pattern noise in image data |
Families Citing this family (28)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2426881B (en) * | 2005-06-01 | 2010-12-29 | Snell & Wilcox Ltd | Method and apparatus for spatial interpolation of colour images |
US7924483B2 (en) * | 2006-03-06 | 2011-04-12 | Smith Scott T | Fused multi-array color image sensor |
KR100780932B1 (en) * | 2006-05-30 | 2007-11-30 | 엠텍비젼 주식회사 | Color interpolation method and device |
US8417759B2 (en) * | 2007-04-09 | 2013-04-09 | DigitalOptics Corporation Europe Limited | Efficient implementations of kernel computations |
US20080260291A1 (en) * | 2007-04-17 | 2008-10-23 | Nokia Corporation | Image downscaling by binning |
US20090077359A1 (en) * | 2007-09-18 | 2009-03-19 | Hari Chakravarthula | Architecture re-utilizing computational blocks for processing of heterogeneous data streams |
US20090092338A1 (en) * | 2007-10-05 | 2009-04-09 | Jeffrey Matthew Achong | Method And Apparatus For Determining The Direction of Color Dependency Interpolating In Order To Generate Missing Colors In A Color Filter Array |
US8270713B2 (en) * | 2007-10-16 | 2012-09-18 | Aptina Imaging Corporation | Method and apparatus providing hardware-efficient demosaicing of image data |
US8086034B2 (en) | 2008-02-05 | 2011-12-27 | Aptina Imaging Corporation | System and method for reducing color artifacts in digital images |
JP5076186B2 (en) * | 2008-10-23 | 2012-11-21 | 株式会社メガチップス | Image enlargement method |
US8073246B2 (en) * | 2008-11-07 | 2011-12-06 | Omnivision Technologies, Inc. | Modifying color and panchromatic channel CFA image |
US8035698B2 (en) * | 2009-01-21 | 2011-10-11 | Seiko Epson Corporation | Joint automatic demosaicking and white balancing |
JP5326943B2 (en) * | 2009-08-31 | 2013-10-30 | ソニー株式会社 | Image processing apparatus, image processing method, and program |
JP2012239103A (en) * | 2011-05-13 | 2012-12-06 | Sony Corp | Image processing device, image processing method, and program |
TW201305965A (en) * | 2011-07-18 | 2013-02-01 | Novatek Microelectronics Corp | Image processing method |
KR101327227B1 (en) | 2012-02-21 | 2013-11-11 | 중앙대학교 산학협력단 | Apparatus and method for noise reduction considering image structure |
CN102938843B (en) * | 2012-11-22 | 2014-12-03 | 华为技术有限公司 | Image processing method, image processing device and imaging device |
JP2015065270A (en) * | 2013-09-25 | 2015-04-09 | ソニー株式会社 | Solid state image pickup device and manufacturing method of the same, and electronic apparatus |
JP6415094B2 (en) * | 2014-04-25 | 2018-10-31 | キヤノン株式会社 | Image processing apparatus, imaging apparatus, image processing method, and program |
KR102224742B1 (en) | 2014-06-10 | 2021-03-09 | 삼성디스플레이 주식회사 | Image display method |
JP6478641B2 (en) * | 2015-01-08 | 2019-03-06 | キヤノン株式会社 | Image processing apparatus, information processing method, and program |
KR101660447B1 (en) * | 2015-03-26 | 2016-09-28 | 인천대학교 산학협력단 | Multi directional weighted interpolation method for Bayer pattern CFA demosaicking |
US20180288281A1 (en) * | 2016-01-25 | 2018-10-04 | Hewlett-Packard Development Company, L.P. | Interpolating pixel values |
CN109993693B (en) * | 2017-12-29 | 2023-04-25 | 澜至电子科技(成都)有限公司 | Method and apparatus for interpolating an image |
CN111163301B (en) * | 2018-11-07 | 2021-04-20 | 浙江宇视科技有限公司 | Color adjustment method, device and computer readable storage medium |
CN109829852B (en) * | 2019-01-17 | 2023-04-25 | 浙江大华技术股份有限公司 | Color filter array image processing method, device and storage medium |
CN110517189B (en) * | 2019-08-02 | 2023-03-21 | Oppo广东移动通信有限公司 | Image processing method and apparatus, and storage medium |
KR102462804B1 (en) * | 2021-06-28 | 2022-11-04 | 주식회사 아이닉스 | Lens distortion correction apparatus based on bayer pattern image and correction method therefor |
Family Cites Families (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3971065A (en) * | 1975-03-05 | 1976-07-20 | Eastman Kodak Company | Color imaging array |
US5852475A (en) * | 1995-06-06 | 1998-12-22 | Compression Labs, Inc. | Transform artifact reduction process |
US6181376B1 (en) * | 1997-10-14 | 2001-01-30 | Intel Corporation | Method of determining missing color values for pixels in a color filter array |
JPH11122626A (en) * | 1997-10-17 | 1999-04-30 | Nikon Corp | Image processing method, system and record medium recording image processing program |
US6614474B1 (en) * | 1998-08-27 | 2003-09-02 | Polycom, Inc. | Electronic pan tilt zoom video camera with adaptive edge sharpening filter |
US6674903B1 (en) * | 1998-10-05 | 2004-01-06 | Agfa-Gevaert | Method for smoothing staircase effect in enlarged low resolution images |
US6642962B1 (en) * | 1999-09-01 | 2003-11-04 | Neomagic Corp. | Merged pipeline for color interpolation and edge enhancement of digital images |
US6847396B1 (en) * | 2000-04-26 | 2005-01-25 | Sunplus Technology Co., Ltd. | Interpolation method for producing full color images in using a single-chip color sensor |
US6707937B1 (en) * | 2000-07-14 | 2004-03-16 | Agilent Technologies, Inc. | Interpolation of edge portions of a digital image |
US6978050B2 (en) * | 2001-07-18 | 2005-12-20 | Hewlett-Packard Development Company, L.P. | Electronic image color plane reconstruction |
KR100396898B1 (en) * | 2001-09-13 | 2003-09-02 | 삼성전자주식회사 | A device and method for processing the output data from image sensor |
WO2003077542A1 (en) * | 2002-03-11 | 2003-09-18 | Koninklijke Philips Electronics N.V. | Image display apparatus |
US7542082B2 (en) * | 2004-03-30 | 2009-06-02 | Canon Kabushiki Kaisha | Method and apparatus for correcting a defective pixel |
KR100587143B1 (en) * | 2004-05-06 | 2006-06-08 | 매그나칩 반도체 유한회사 | Method for edge detection of image signal |
JP4333997B2 (en) * | 2004-08-24 | 2009-09-16 | シャープ株式会社 | Image processing apparatus, photographing apparatus, image processing method, image processing program, and recording medium |
-
2005
- 2005-12-29 US US11/319,398 patent/US7830426B2/en active Active
-
2006
- 2006-12-19 CN CN2006800519591A patent/CN101336438B/en active Active
- 2006-12-19 JP JP2008548595A patent/JP4893973B2/en active Active
- 2006-12-19 KR KR1020087017803A patent/KR101007754B1/en active IP Right Grant
- 2006-12-19 WO PCT/US2006/048200 patent/WO2007078912A1/en active Application Filing
- 2006-12-19 EP EP06845702A patent/EP1969558B1/en active Active
- 2006-12-19 AT AT06845702T patent/ATE520100T1/en not_active IP Right Cessation
- 2006-12-29 TW TW095149970A patent/TWI337499B/en active
-
2010
- 2010-09-24 US US12/889,527 patent/US8212900B2/en active Active
Non-Patent Citations (6)
Title |
---|
ALEKSIC M ET AL: "Local correlation based CFA interpolation", ELECTRICAL AND COMPUTER ENGINEERING, 2004. CANADIAN CONFERENCE ON NIAGARA FALLS, ONT., CANADA 2-5 MAY 2004, PISCATAWAY, NJ, USA,IEEE, US, 2 May 2004 (2004-05-02), pages 793 - 796Vol2, XP010733565, ISBN: 0-7803-8253-6 * |
ALTUNBASAK Y ET AL: "Demosaicking: color filter array interpolation [exploring the imaging process and the correlations among three color planes in single-chip digital cameras]", IEEE SIGNAL PROCESSING MAGAZINE, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 22, no. 1, January 2005 (2005-01-01), pages 44 - 54, XP011128684, ISSN: 1053-5888 * |
CHANG E ET AL: "Color filter array recovery using a threshold-based variable number of gradients", PROCEEDINGS OF THE SPIE, SPIE, BELLINGHAM, VA, US, vol. 3650, 27 January 1999 (1999-01-27), pages 36 - 43, XP002310496, ISSN: 0277-786X * |
KEHTARNAVAZ N ET AL: "COLOR FILTER ARRAY INTERPOLATION USING COLOR CORRELATIONS AND DIRECTIONAL DERIVATIVES", JOURNAL OF ELECTRONIC IMAGING, SPIE / IS & T, US, vol. 12, no. 4, October 2003 (2003-10-01), pages 621 - 632, XP001178695, ISSN: 1017-9909 * |
RAMANATH R ET AL: "DEMOSAICKING METHODS FOR BAYER COLOR ARRAYS", JOURNAL OF ELECTRONIC IMAGING, SPIE / IS & T, US, vol. 11, no. 3, July 2002 (2002-07-01), pages 306 - 315, XP001122626, ISSN: 1017-9909 * |
XIAOMENG WANG ET AL: "Demosaicing with Improved Edge Direction Detection", CIRCUITS AND SYSTEMS, 2005. ISCAS 2005. IEEE INTERNATIONAL SYMPOSIUM ON KOBE, JAPAN 23-26 MAY 2005, PISCATAWAY, NJ, USA,IEEE, 23 May 2005 (2005-05-23), pages 2048 - 2051, XP010815985, ISBN: 0-7803-8834-8 * |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8330772B2 (en) | 2009-10-20 | 2012-12-11 | Apple Inc. | Image signal processor front-end image data processing system and method |
US8358319B2 (en) | 2009-10-20 | 2013-01-22 | Apple Inc. | System and method for processing image data using an image processing pipeline of an image signal processor |
US8593483B2 (en) | 2009-10-20 | 2013-11-26 | Apple Inc. | Temporal filtering techniques for image signal processing |
US8638342B2 (en) | 2009-10-20 | 2014-01-28 | Apple Inc. | System and method for demosaicing image data using weighted gradients |
US8471932B2 (en) | 2010-09-30 | 2013-06-25 | Apple Inc. | Spatial filtering for image signal processing |
US9077943B2 (en) | 2012-05-31 | 2015-07-07 | Apple Inc. | Local image statistics collection |
US9131196B2 (en) | 2012-05-31 | 2015-09-08 | Apple Inc. | Systems and methods for defective pixel correction with neighboring pixels |
US8917336B2 (en) | 2012-05-31 | 2014-12-23 | Apple Inc. | Image signal processing involving geometric distortion correction |
US8953882B2 (en) | 2012-05-31 | 2015-02-10 | Apple Inc. | Systems and methods for determining noise statistics of image data |
US9014504B2 (en) | 2012-05-31 | 2015-04-21 | Apple Inc. | Systems and methods for highlight recovery in an image signal processor |
US9025867B2 (en) | 2012-05-31 | 2015-05-05 | Apple Inc. | Systems and methods for YCC image processing |
US9031319B2 (en) | 2012-05-31 | 2015-05-12 | Apple Inc. | Systems and methods for luma sharpening |
US8817120B2 (en) | 2012-05-31 | 2014-08-26 | Apple Inc. | Systems and methods for collecting fixed pattern noise statistics of image data |
US9105078B2 (en) | 2012-05-31 | 2015-08-11 | Apple Inc. | Systems and methods for local tone mapping |
US8872946B2 (en) | 2012-05-31 | 2014-10-28 | Apple Inc. | Systems and methods for raw image processing |
US9142012B2 (en) | 2012-05-31 | 2015-09-22 | Apple Inc. | Systems and methods for chroma noise reduction |
US9317930B2 (en) | 2012-05-31 | 2016-04-19 | Apple Inc. | Systems and methods for statistics collection using pixel mask |
US9332239B2 (en) | 2012-05-31 | 2016-05-03 | Apple Inc. | Systems and methods for RGB image processing |
US9342858B2 (en) | 2012-05-31 | 2016-05-17 | Apple Inc. | Systems and methods for statistics collection using clipped pixel tracking |
US9710896B2 (en) | 2012-05-31 | 2017-07-18 | Apple Inc. | Systems and methods for chroma noise reduction |
US9741099B2 (en) | 2012-05-31 | 2017-08-22 | Apple Inc. | Systems and methods for local tone mapping |
US9743057B2 (en) | 2012-05-31 | 2017-08-22 | Apple Inc. | Systems and methods for lens shading correction |
US11089247B2 (en) | 2012-05-31 | 2021-08-10 | Apple Inc. | Systems and method for reducing fixed pattern noise in image data |
US11689826B2 (en) | 2012-05-31 | 2023-06-27 | Apple Inc. | Systems and method for reducing fixed pattern noise in image data |
Also Published As
Publication number | Publication date |
---|---|
JP2009522864A (en) | 2009-06-11 |
US20110032397A1 (en) | 2011-02-10 |
KR20080078076A (en) | 2008-08-26 |
TW200735652A (en) | 2007-09-16 |
CN101336438A (en) | 2008-12-31 |
US7830426B2 (en) | 2010-11-09 |
ATE520100T1 (en) | 2011-08-15 |
EP1969558A1 (en) | 2008-09-17 |
JP4893973B2 (en) | 2012-03-07 |
US20070153106A1 (en) | 2007-07-05 |
EP1969558B1 (en) | 2011-08-10 |
CN101336438B (en) | 2011-11-09 |
KR101007754B1 (en) | 2011-01-14 |
US8212900B2 (en) | 2012-07-03 |
TWI337499B (en) | 2011-02-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1969558B1 (en) | Method and apparatus providing color interpolation in color filter arrays using edge detection and correction terms | |
US7456881B2 (en) | Method and apparatus for producing Bayer color mosaic interpolation for imagers | |
US10021358B2 (en) | Imaging apparatus, imaging system, and signal processing method | |
KR101130023B1 (en) | Solid-state image sensing device | |
US8164651B2 (en) | Concentric exposure sequence for image sensor | |
US8339489B2 (en) | Image photographing apparatus, method and medium with stack-type image sensor, complementary color filter, and white filter | |
KR101012537B1 (en) | Solid-state image sensor | |
US6781626B1 (en) | System and method of color interpolation | |
US7876363B2 (en) | Methods, systems and apparatuses for high-quality green imbalance compensation in images | |
US20080151083A1 (en) | Color filter pattern for color filter arrays including a demosaicking algorithm | |
US20070133893A1 (en) | Method and apparatus for image noise reduction | |
US20080055436A1 (en) | Method, imager and system providing paired-bayer color filter array and interlaced readout | |
JP2003298039A (en) | Digital image sensor having dichroic photodetector | |
CA2443658C (en) | Method and apparatus for sensing and interpolating color image data | |
CN207251823U (en) | Imaging device and imaging system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2008548595 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2006845702 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020087017803 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 200680051959.1 Country of ref document: CN |