CN105323568A - Color reconstruction method of color filter array in digital camera - Google Patents
Color reconstruction method of color filter array in digital camera Download PDFInfo
- Publication number
- CN105323568A CN105323568A CN201410368657.3A CN201410368657A CN105323568A CN 105323568 A CN105323568 A CN 105323568A CN 201410368657 A CN201410368657 A CN 201410368657A CN 105323568 A CN105323568 A CN 105323568A
- Authority
- CN
- China
- Prior art keywords
- pixel
- value
- primary
- rgb
- average
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Landscapes
- Color Television Image Signal Generators (AREA)
Abstract
The invention discloses a color reconstruction method of a color filter array in a digital camera. The method comprises the steps that an edge determining method is used to interpolate a default G primary color value for R and B pixels, and at the same time the R primary color value of the B pixel is preliminarily estimated; the Cr value of the R pixel and the Cb value of the B pixel are approximately estimated; Cb and Cr values of G pixels are interpolated; the Cb value of the R pixel and the Cr value of the B pixel are interpolated; a conversion formula which uses the mean of Cb and Cr values of left and right G pixels to represent a YCbCr chroma space and a RGB three-primary-color space is used to inverse the default primary color values. According to the invention, a digital photo is clear; compared with RGB three-primary-color interpolation, CbCr interpolation of a digital image pixel is more approximate to the true value and has a better effect; the edges of an image are clear; noise is removed; the processing speed is fast.
Description
Technical field
The present invention relates to the color reconstruction method of colour filter array in a kind of digital camera, for digital imaging technology field.
Background technology
Common digital camera and the image device of Digital Video are all charge-coupled device (CCD)s, and it is the device using charge packet as picture signal carrier, utilize photon to inspire electronics in monocrystalline silicon---the right light sensitive effect in hole realizes opto-electronic conversion.Charge coupled device (CCD) is a mass of photoelectric conversion unit small one by one, i.e. the pixel one by one of corresponding digital picture.Digital camera activates CCD by shutter, light signal is converted into the signal of telecommunication, each photoelectric conversion unit changes into voltage signal the intensity of illumination that it accepts, be digital signal by transform electrical signals again, according to array order, each value is preserved in memory, so just obtain original digital picture.
The common various coloured light of occurring in nature all can be matched by different proportion by red (R), green (G), blue (B) three kinds of coloured light and form.Charge coupled device (CCD) itself can only perception light intensity, in order to obtain coloured image, CCD device employs color filter.Each photoelectric conversion unit comprises the micro optical filter (or R or G or B) of a different colours, and they press order array as mosaic of G-R-G-B.Each like this photoelectric conversion unit just can obtain the response value of often kind of coloured light, and therefore each pixel of original image only has a three primary colors component: or red R or green G or blue B
.but each pixel of real color digital image needs R, G, B tri-components, and this just needs interpolation algorithm, from raw information interpolation go out each pixel the color component that lacks.
Color interpolation is the key factor affecting last picture quality.Find by literature search, Chinese Patent Application No.: 02116750.8, patent name: the color reconstruction method of colour filter array in digital camera.The patent proposes the interpolation algorithm in a kind of digital camera, its feature is: the green component of green pixel point is represented by itself, and its red component and blue component are represented by the average of its average of upper and lower 2 or 2, left and right; The red component of red pixel point is represented by itself, its blue component by upper left, lower-left, the mean value of the amount of blue of 4 that upper right is adjacent with bottom right represents; Its green component by upper, under, the intermediate value of the amount of green color that left and right is 4 represents; The blue component of blue picture element point is represented by itself, its red component by upper left, lower-left, the mean value of the amount of red of 4 that upper right is adjacent with bottom right represents; Its green component by upper, under, the intermediate value of the amount of green color that left and right is 4 represents; Median computation methods is: first sorted by size by the component value of four points, then gets the intermediate value of mean value as this four number of two values in the middle of it.Its weak point is: the gradient of the edge's primary color values in image is very large, uses the average of neighboring pixel and intermediate value to carry out the fuzzy edge of interpolation, the edge of image also can be caused to occur light and dark sawtooth.
Summary of the invention
For solving above-mentioned existing shortcoming, main purpose of the present invention is to provide the color reconstruction method of colour filter array in a kind of digital camera of practicality, makes digital photograph more clear, abates the noise, and can also ensure processing speed faster.
For reaching above-described object, in a kind of digital camera of the present invention, the color reconstruction method of colour filter array takes following technical scheme:
In digital camera, a color reconstruction method for colour filter array, is characterized in that, comprise step as follows
:
The first step, with edge determining method, default G primary color values is inserted out to R pixel and B pixel, the simultaneously R primary color values of B pixel according to a preliminary estimate, namely compare according to the absolute difference of R pixel or the upper and lower G pixel of B pixel and left and right G pixel value difference absolute value, if the former is greater than the latter, think that this place exists horizontal edge, the G primary colours of R pixel or B pixel represent by the average of upper and lower G pixel, the R primary color values of B pixel temporarily represents by its upper left corner R pixel and the average of upper right corner R pixel roughly, on the contrary, think that this place exists vertical edge, represent by the average of left and right G pixel, the R primary color values of B pixel temporarily represents by its upper left corner R pixel and the average of lower left corner R pixel
,
Second step, the Cr value of approximate evaluation R pixel and the Cb value of B pixel, namely according to the conversion formula in YCbCr chrominance space and RGB three primary colors space, the G primary color values original R primary colours and the first step of R pixel obtained substitutes into its Cr value of conversion formula approximate calculation in YCbCr chrominance space and RGB three primary colors space, ignore the impact of B primary colours herein, the G primary color values original B primary color values and the first step of B pixel obtained substitutes into the conversion formula in YCbCr chrominance space and RGB three primary colors space, also its preliminary R primary colours estimated value is substituted into, its Cb value of approximate calculation;
3rd step, interpolation is carried out to Cb, Cr value of G pixel, in even number line, the Cr value of G pixel represents by the average of the Cr value of left and right R pixel, the Cb value average of the Cb value of upper and lower B pixel represents, in odd-numbered line, the Cr value of G pixel represents by the average of the Cr value of upper and lower R pixel, and the Cb value average of the Cb value of left and right B pixel represents;
4th step, carries out interpolation to the Cb value of R pixel and the Cr value of B pixel, all represents by the average of Cb, Cr value of its left and right G pixel;
5th step, according to the Cb value and the Cr value that obtain each pixel above, pushes away their default primary color values according to the conversion formula in YCbCr chrominance space and RGB three primary colors space is counter.
The conversion formula in described YCbCr chrominance space and RGB three primary colors space is:
Y0.29900.58700R
Cb=-0.1687-0.3313128G
Cr0.5000-0.4187128B
I0011
Described YCbCr chrominance space, it is the color representation based on human eye vision reaction, Y represents lightness, it is the sensation that colour acts on the light levels that human eye causes, represent colourity, the classification of reflection color, be called chroma blue and red color, lightness Y is created by flesh RGB input signal, method is superimposed together by the specific part of rgb signal, and Cr reflects the difference between RGB input signal RED sector and signal angle value, Cb reflection be RGB input signal blue portion and the difference together of rgb signal brightness value.
Adopt the present invention of as above technical scheme, there is following beneficial effect:
Make digital photograph more clear, compared with RGB three primary colors interpolation, carrying out interpolation to the CbCr of Pixel of Digital Image will more approach its actual value, has better effect.The edge clear of image can be made, stress release treatment, and processing speed faster can also be ensured.
Accompanying drawing explanation
Fig. 1 is the distribution map of the micro optical filter of CCD.
Fig. 2 is 6*16 filter arrays figure.
Fig. 3 is color interpolation flow chart of the present invention.
Embodiment
In order to further illustrate the present invention, be described further below in conjunction with accompanying drawing:
In the array of charge coupled device (CCD) colo(u)r filter, red, green, blue three-colour filter distributed inerratic, wherein Red lightscreening plate and blue color filter only account for whole 1/4th respectively, and green color filter accounts for whole 1/2nd, its distribution as shown in Figure 1.The original digital picture utilizing CCD to obtain as shown in Figure 2.
The view picture original image obtained for CCD implements the difference approach that the present invention relates to, realize by loop structure, handling object using six row pixels as a circulation, from zero row to fifth line, to middle 2,3 liang of row pixel value differences, complete a circulation, down move two row, carry out next one circulation, as shown in Figure 3, treatment step of the present invention is as follows:
(1) insert out default G primary color values by the B pixel of edge determining method to 1,3 row, simultaneously its R primary color values according to a preliminary estimate, default G primary color values is inserted out to the R pixel of 2,4 row.Namely compare according to the absolute difference of R pixel or the upper and lower G pixel of B pixel and left and right 6 pixel value difference absolute value, if the former is greater than the latter, think that this place exists horizontal edge, the G primary color values of R pixel or B pixel represents by the average of upper and lower G pixel, the R primary color values of B pixel temporarily represents by its upper left corner R pixel and the average of upper right corner R pixel roughly, on the contrary, think that this place exists vertical edge, represent by the average of left and right G pixel, the R primary color values of B pixel temporarily represents by its upper left corner R pixel and the average of lower left corner R pixel roughly.
(2) the Cr value and 1 of approximate evaluation 2,4 row R pixel, the Cb value of 3 row B pixels.Namely according to the conversion formula in YCbCr chrominance space and RGB three primary colors space, the G primary color values original R primary colours and the first step of R pixel obtained substitutes into its Cb value of conversion formula approximate calculation in chrominance space and RGB three primary colors space, ignore the impact of B primary colours herein, the G primary color values that the original B primary color values of B pixel, the first step are obtained and its conversion formula being worth substitution chrominance space and RGB three primary colors space compared with primary colours according to a preliminary estimate, its Cb value of approximate calculation.
(3) interpolation is carried out to Cb, Cr value of 2,3 row G pixels.For the 2nd row, the Cr value of G pixel represents by the average of the Cr value of left and right R pixel, and the Cb value average of the Cb value of upper and lower B pixel represents, for the 3rd row, the Cr value of G pixel represents by the average of the Cr value of upper and lower R pixel, and the Cb value average of the Cb value of left and right B pixel represents;
(4) interpolation is carried out to the Cb value of R pixel and the Cr value of B pixel, all represent by the average of Cb, Cr value of its left and right G pixel;
(5) according to the Cb value and the Cr value that to obtain pixel above, their default primary color values are pushed away according to the conversion formula in YCbCr chrominance space and RGB three primary colors space is counter; Lower movement two row, if be last column, end loop, otherwise returns step (1).
Be exactly more than the specific embodiment of interpolation algorithm of the present invention, can realize fast in the digital signal processor of digital camera (DSP), not only details is clear for institute's image that obtains, and can reduce noise.
Claims (3)
1. the color reconstruction method of colour filter array in digital camera, is characterized in that, comprise step as follows:
The first step, with edge determining method, default G primary color values is inserted out to R pixel and B pixel, the simultaneously R primary color values of B pixel according to a preliminary estimate, namely compare according to the absolute difference of R pixel or the upper and lower G pixel of B pixel and left and right G pixel value difference absolute value, if the former is greater than the latter, think that this place exists horizontal edge, the G primary colours of R pixel or B pixel represent by the average of upper and lower G pixel, the R primary color values of B pixel temporarily represents by its upper left corner R pixel and the average of upper right corner R pixel roughly, on the contrary, think that this place exists vertical edge, represent by the average of left and right G pixel, the R primary color values of B pixel temporarily represents by its upper left corner R pixel and the average of lower left corner R pixel
,
Second step, the Cr value of approximate evaluation R pixel and the Cb value of B pixel, namely according to the conversion formula in YCbCr chrominance space and RGB three primary colors space, the G primary color values original R primary colours and the first step of R pixel obtained substitutes into its Cr value of conversion formula approximate calculation in YCbCr chrominance space and RGB three primary colors space, ignore the impact of B primary colours herein, the G primary color values original B primary color values and the first step of B pixel obtained substitutes into the conversion formula in YCbCr chrominance space and RGB three primary colors space, also its preliminary R primary colours estimated value is substituted into, its Cb value of approximate calculation;
3rd step, interpolation is carried out to Cb, Cr value of G pixel, in even number line, the Cr value of G pixel represents by the average of the Cr value of left and right R pixel, the Cb value average of the Cb value of upper and lower B pixel represents, in odd-numbered line, the Cr value of G pixel represents by the average of the Cr value of upper and lower R pixel, and the Cb value average of the Cb value of left and right B pixel represents;
4th step, carries out interpolation to the Cb value of R pixel and the Cr value of B pixel, all represents by the average of Cb, Cr value of its left and right G pixel;
5th step, according to the Cb value and the Cr value that obtain each pixel above, pushes away their default primary color values according to the conversion formula in YCbCr chrominance space and RGB three primary colors space is counter.
2. the color reconstruction method of colour filter array in a kind of digital camera according to claim 1, it is characterized in that, the conversion formula in described YCbCr chrominance space and RGB three primary colors space is:
Y0.29900.58700R
Cb=-0.1687-0.3313128G
Cr0.5000-0.4187128B
I0011。
3. the color reconstruction method of colour filter array in a kind of digital camera according to claim 1, it is characterized in that, described YCbCr chrominance space, it is the color representation based on human eye vision reaction, Y represents lightness, it is the sensation that colour acts on the light levels that human eye causes, represent colourity, the classification of reflection color, be called chroma blue and red color, lightness Y is created by flesh RGB input signal, method is superimposed together by the specific part of rgb signal, and Cr reflects the difference between RGB input signal RED sector and signal angle value, Cb reflection be RGB input signal blue portion and the difference together of rgb signal brightness value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410368657.3A CN105323568A (en) | 2014-07-30 | 2014-07-30 | Color reconstruction method of color filter array in digital camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410368657.3A CN105323568A (en) | 2014-07-30 | 2014-07-30 | Color reconstruction method of color filter array in digital camera |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105323568A true CN105323568A (en) | 2016-02-10 |
Family
ID=55250045
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410368657.3A Withdrawn CN105323568A (en) | 2014-07-30 | 2014-07-30 | Color reconstruction method of color filter array in digital camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105323568A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113824936A (en) * | 2021-09-23 | 2021-12-21 | 合肥埃科光电科技有限公司 | Color interpolation method, device and equipment for color filter array line scanning camera |
| CN113965735A (en) * | 2020-07-21 | 2022-01-21 | 瑞昱半导体股份有限公司 | Image decolouring band judging method and its image decolouring band judging circuit |
-
2014
- 2014-07-30 CN CN201410368657.3A patent/CN105323568A/en not_active Withdrawn
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113965735A (en) * | 2020-07-21 | 2022-01-21 | 瑞昱半导体股份有限公司 | Image decolouring band judging method and its image decolouring band judging circuit |
| CN113965735B (en) * | 2020-07-21 | 2024-03-01 | 瑞昱半导体股份有限公司 | Image color-removing band judging method and image color-removing band judging circuit |
| CN113824936A (en) * | 2021-09-23 | 2021-12-21 | 合肥埃科光电科技有限公司 | Color interpolation method, device and equipment for color filter array line scanning camera |
| CN113824936B (en) * | 2021-09-23 | 2024-02-09 | 合肥埃科光电科技股份有限公司 | Color interpolation method, device and equipment for color filter array line scanning camera |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR100982685B1 (en) | System and method for a high performance color filter mosaic array | |
| US10645268B2 (en) | Image processing method and apparatus of terminal, and terminal | |
| WO2021208593A1 (en) | High dynamic range image processing system and method, electronic device, and storage medium | |
| WO2021196554A1 (en) | Image sensor, processing system and method, electronic device, and storage medium | |
| US7400332B2 (en) | Hexagonal color pixel structure with white pixels | |
| US9482797B2 (en) | Pixel array, camera using the same and color processing method based on the pixel array | |
| US8723995B2 (en) | Extended dynamic range in color imagers | |
| EP2664153B1 (en) | Imaging system using a lens unit with longitudinal chromatic aberrations and method of operating | |
| US8749672B2 (en) | Digital camera having a multi-spectral imaging device | |
| WO2021212763A1 (en) | High-dynamic-range image processing system and method, electronic device and readable storage medium | |
| US8724894B1 (en) | Colorization of digital imagery | |
| US11632525B2 (en) | Image processing method and filter array including wideband filter elements and narrowband filter elements | |
| CN103327220B (en) | With green channel for the denoising method guided on low-light (level) Bayer image | |
| CN102457682A (en) | Camera system and imaging method using multiple lens and aperture units | |
| WO2023016146A1 (en) | Image sensor, image collection apparatus, image processing method, and image processor | |
| CN104394391B (en) | The image processing method of the dot matrix image of collected by camera and system | |
| JP2015194567A (en) | display device | |
| CN103259960B (en) | The interpolation method of data and device, image output method and device | |
| CN105323568A (en) | Color reconstruction method of color filter array in digital camera | |
| CN103546729B (en) | Gray-color dual-mode TDI-CMOS image sensor and control method | |
| CN108513112A (en) | Color filter array and image processing method | |
| CN104754321A (en) | Camera array system | |
| CN1252983C (en) | Method of color interpolation in digital camera | |
| CN203722755U (en) | Grey and color dual-mode TDI-CMOS image sensor | |
| CN110324541A (en) | The filtration combined denoising interpolation method of one kind and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C04 | Withdrawal of patent application after publication (patent law 2001) | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20160210 |