WO2019090909A1 - Method and device for improving contrast of bayer image - Google Patents

Abstract

The present invention provides a method and device for improving the contrast of a Bayer image. The method comprises: respectively obtaining dark primary color image data of a bottom portion, a middle portion, and a top portion of a pixel unit in the current Bayer image data; managing the obtained dark primary color image data in time-sharing fashion, obtaining integral values of a top left region TL and a top right region TR by means of the dark primary color image data of the top portion, obtaining integral values of a bottom left region BL and a bottom right region BR by means of the dark primary color image data of the bottom portion, and forming an integral image; performing comprehensive processing on the integral image and the integral values of the top left region TL, the top right region TR, the bottom left region BL, and the bottom right region BR, and then preforming bilateral filtering processing to obtain filtered dark primary color image data; and performing computational processing on the filtered dark primary image data to obtain Bayer image data having improved contrast. The method simplifies an intermediate processing procedure and reduces processing delays while ensuring the contrast and detailed information of an image.

Description

Method and device for improving Bayer image contrast

[Technical Field]

The present invention relates to the field of image processing technologies, and in particular, to a method and apparatus for improving Bayer image contrast.

 【Background technique】

In the process of image capture and video capture, when night vision or backlight is often encountered, the image obtained at this time has low contrast and blurred details, which brings inconvenience to subsequent image analysis work. The use of image enhancement technology can effectively improve the performance of systems such as video capture, image recognition, and dynamic recognition under low illumination conditions.

For color images, you need to collect a variety of basic colors, such as RGB three colors, the easiest way is to use the filter method, the red filter through the red wavelength, the green filter through the green wavelength, The blue filter passes through the blue wavelength. If you want to capture the three basic colors of RGB, you need three filters, which is expensive and not easy to manufacture, because all three filters must ensure that each pixel is aligned. When using the bayer format, this problem is solved very well. The bayer image is a different color set on a filter and is found by analyzing the perception of color by the human eye. Bayer mode is a color mode that is widely used in CCD and CMOS cameras. The contrast enhancement method for Bayer image data in the prior art mainly comprises the following steps: first performing interpolation processing on Bayer image data to obtain interpolated RGB image data, and then acquiring dark primary color image data based on RGB image data; for dark primary color image The integral map is calculated, and the acquired dark primary color image is bilaterally filtered by the integral map to obtain the filtered dark primary color image data; the histogram statistics are performed on the RGB image data and the filtered dark primary color image data to obtain R/ G/B (three component values of pixel points R, G, and B) and histogram of dark primary image data; at the same time, cumulative statistics are performed on the filtered dark primary image data to obtain the sum of dark primary image data in the filtering window; G/B, dark primary image data histogram, total analysis of dark primary image data in the filtering window, comprehensive analysis, get gain value, dark channel mean and R/G/B corresponding atmospheric light value; use each frame to update the gain value And the atmospheric light value, the dark channel mean value and the filtered dark primary color image data corresponding to R/G/B, and the original Bayer image data Calculation processing, Bayer finally obtained image data to enhance the contrast.

In the prior art, it is necessary to perform buffer processing on Bayer image data, bilaterally filtered front/back dark primary color image data, and integral map, and requires memory access multiple times while occupying memory; in addition, in the original method, in addition to statistics/maintenance R /G/B histogram, also need to do the calculation process of multiple intermediate processes. Therefore, in the existing contrast enhancement method for Bayer image data, both the graphics processing and the digital signal processing implementation meet the real-time requirements, and even the processing delay reaches several seconds.

Therefore, there is a need for an improved Bayer image contrast scheme that can improve the processing delay while ensuring image contrast and detail information.

 [Summary of the Invention]

The technical problem to be solved by the present invention is to provide a method and apparatus for improving the contrast of a Bayer image, which can simplify the intermediate processing process and reduce the processing delay while ensuring the contrast and detail information of the image.

To solve the above technical problems, the present invention provides a method for improving the contrast of a Bayer image, comprising:

Interpolating current Bayer image data to obtain interpolated RGB image data;

Obtaining dark primary color image data of the bottom, middle, and top of the pixel unit according to the interpolated RGB image data;

The acquired dark primary color image data is time-divisionally managed, and the integrated values of the left top region TL and the right top region TR are obtained by the dark primary color image data of the top, and the left bottom region BL and the right are obtained by the dark primary color image data of the bottom. The integral value of the bottom region BR and form an integral map;

The integrated values of the integral map, the left top region TL, the right top region TR, the left bottom region BL, and the right bottom region BR are integrated to obtain a sum of dark primary color image data in the filtered window. Performing bilateral filtering processing on the sum of the dark primary color image data in the filtering window to obtain filtered dark primary color image data;

Performing histogram statistics on the RGB image data and the filtered dark primary color image to obtain an R/G/B image data cumulative histogram, a dark primary color image data histogram, and a dark primary color image data cumulative total value;

Integrating the R/G/B image data accumulation histogram, the dark primary color image data histogram, and the dark primary color image data cumulative total value to obtain a gain value and an atmosphere corresponding to the R/G/B image data. Light value

The gain value obtained by the Bayer image data and the atmospheric light value corresponding to the R/G/B image data, the filtered dark primary color image, and the Bayer image data are integrated to obtain Bayer image data with contrast enhancement.

In order to solve the above technical problem, the present invention also provides a method for improving the contrast of a Bayer image, comprising: a method for improving the contrast of a Bayer image, comprising respectively acquiring dark primary color image data of the bottom, middle and top of the pixel unit in the current Bayer image data. The obtained dark primary color image data is time-divisionally managed, and the integrated values of the left top region TL and the right top region TR are obtained by the top dark primary image data of the top, and the left bottom region BL and the left bottom region BL are obtained by the bottom dark primary image data of the bottom An integral value of the right bottom region BR, and forming an integral map; an integral value of the integral map, the left top region TL, the right top region TR, the left bottom region BL, and the right bottom region BR After the comprehensive processing, the sum of the dark primary color image data in the filtering window is obtained, and the sum of the dark primary color image data in the filtering window is bilaterally processed to obtain the filtered dark primary color image data; and the filtered dark primary color image data is calculated and processed. , get the Bayer image data after the contrast is improved.

In order to solve the above problems, the present invention also provides an apparatus for improving the contrast of a Bayer image, comprising: a storage manager for realizing buffering of Bayer image line data in a bilateral filtering window; a bottom running module, the bottom running module And configured to obtain bottom dark primary image data of the pixel unit in the Bayer image data, and calculate an integrated value of the left bottom region BL and the right bottom region BR according to the bottom dark primary image data; the top running module, the top running module Obtaining top dark primary color image data of the pixel unit in the Bayer image data, and calculating an integrated value of the left top region TL and the right top region TR according to the top dark primary color image data; the middle running module, the middle running module is used to Forming an integral map, and integrating the integrated values of the integral map, the left top region TL, the right top region TR, the left bottom region BL, and the right bottom region BR to obtain a dark primary color in the filtering window Sum of image data, bilaterally filtering the sum of dark primary color image data in the filtering window to obtain filtered Primary color image data, the image data on dark after having calculated using the filtering process to obtain the image data Bayer enhance the contrast.

The invention has the beneficial effects that the method and the device for improving the contrast of the Bayer image are proposed differently from the prior art. The method and the device of the invention adopt a pipeline and data parallel processing structure, while ensuring the contrast and detail information of the image. Simplify the intermediate processing process, reduce the processing delay, and improve the processing of Bayer images. Further, the method and device of the present invention converts the integral map required for bilateral filtering into an integral line mode for management, optimizes the dark state image, suppresses the over-bright area, and improves the contrast of the entire image.

 [Description of the Drawings]

1 is a schematic flow chart of an embodiment of a method for improving Bayer image contrast according to the present invention;

2 is a schematic diagram of integration of regions of dark primary color image data of the present invention;

3 is a schematic diagram of an integral line of dark primary color image data of the present invention;

4 is a schematic diagram showing the calculation principle of the sum of the dark primary color image data in the filtering window of the present invention;

FIG. 5 is a schematic structural diagram of an apparatus for improving Bayer image contrast according to a first embodiment of the present invention; FIG.

6 is a schematic structural diagram of an apparatus for improving Bayer image contrast according to a second embodiment of the present invention;

Figure 7 is a block diagram showing the structure of an apparatus for improving the contrast of a Bayer image according to a third embodiment of the present invention.

【Detailed ways】

Please refer to FIG. 1. FIG. 1 is a schematic flow chart of an embodiment of a method for improving the contrast of a Bayer image according to the present invention. The method for improving the contrast of a Bayer image in this embodiment includes the following steps:

101: Obtain dark primary color image data of the bottom, middle, and top of the pixel unit in the current Bayer image data, respectively.

Specifically, the current Bayer image data is interpolated to obtain the interpolated RGB image data; and the dark primary color image data of the bottom, middle, and top of the pixel unit is obtained according to the interpolated RGB image data.

In this embodiment, by selecting the minimum value in the RGB image as the dark primary color image data, the bottom running module acquires the bottom dark primary color image data of the pixel unit in the Bayer image data, and the top running module obtains the top of the pixel unit in the Bayer image data. The dark primary image data, the middle running module acquires part of the dark primary image data in the Bayer image data.

102: The dark primary color image data is time-divisionally managed, and the integrated values of the left top region TL and the right top region TR are obtained by the top dark primary image data, and the left bottom region BL and the left bottom region BL are obtained by the dark primary color image data at the bottom. Integrate the values and form an integral map.

Preferably, in order to integrate the synchronization, the top running module and the bottom running module are time-divisionally managed, the bottom running module directly acquires Bayer image data, and the top running module acquires Bayer image data in the cache device.

As shown in FIG. 2, the left top area TL201, the right top area TR202, the left bottom area BL203, and the left bottom area BL204 represent the areas of the respective pixel units to the base point 205, respectively.

In this embodiment, the integral map is implemented by means of integral rows. In a specific embodiment, as shown in FIG. 3, the integration line is used to simulate the implementation of the integral map function, and the left top region TL301 and the right top region TR302 and the left bottom region BL303 and the right are calculated according to the calculation of the cumulative sum in the filter window. The integral value of the bottom area BR304 is managed by time sharing through the top running module and the bottom running module branch; each group of integral lines uses two static memories to simulate the integral value management in a time sharing manner, wherein the static memory A and B correspond to the top integral Lines; while static memories C and D correspond to the bottom of the integral line; when the first line of Bayer image data arrives, the data in static memories A, B, C, and D is cleared; when the first line of Bayer image data arrives, The static memory A will refer to the static memory B as a reference, and simultaneously read the integrated values of (i-1, j-1), (i-1, j) from the static memory B; read from the static memory A (i, The value of j-1); after the integral value formula is calculated, the integral value (i, j) is updated to the corresponding position of the static memory A; when the second line of the Bayer image data arrives, the static memory A is saved at this time. Product of the first line of data Data; static memory B will use static memory A as a reference to calculate the integral value of the current line and update it to static memory B; when the third line of image frame arrives, static memory A and B switch again at this time. The static memory A is again referenced to the static memory B, and the integral map data of the current line is calculated and updated to the static memory A; the static memory C and D are managed in the same manner as the static memories A and B. And so on, until the end of the calculation to get the integral map.

103: The integral value of the dark primary color image data in the filtering window is obtained by integrating the integral values of the integral map, the left top region TL, the right top region TR, the left bottom region BL, and the right bottom region BR, and the dark primary color image data in the filtering window is filtered. The sum is subjected to bilateral filtering processing to obtain filtered dark primary color image data.

Preferably, the calculation principle of the sum of the dark primary image data in the filtering window is as shown in FIG. 4, and I(i, j) is that the information stored in the integral graph is (i, j) pixel unit to base point (0, 0) (Fig. The integral value of the sum of all the pixels in 405), P(i, j) is the pixel unit integral value.

P(0,0), the integral value of the corresponding position of the integral map is:

I(0,0)=P(0,0).

P(0,1), the integral value of the corresponding position of the integral map is:

I(0,1)=P(0,1)+I(0,0).

P(1,0), the integral value of the corresponding position of the integral map is:

I(1,0)=P(1,0)+I(0,0).

P(1,1), the integral value of the corresponding position of the integral map is:

I(1,1)=I(1,0)+I(0,1)-I(0,0)+P(1,1).

Then, for P(i,j), as shown in 401~404 in Fig. 4, the integral value of the corresponding position of the integral map is:

I(i,j)=I(i,j-1)+I(i-1,j)-I(i-1,j-1)+P(i,j).

Through the above result, the sum of the dark primary color image data in the filtering window is obtained by combining the integrated values of the left top region TL and the right bottom region BR, subtracting the integrated values of the right top region TR and the left bottom region BL, and the sum of the dark primary color image data in the window is obtained. Calculation formula:

Sum=BR+TL-TR-BL (1)

(Where, Sum, the sum of the dark primary image data in the window; BR, the integral value of the right bottom region; TL, the integral value of the left top region; TR, the integral value of the right top region; BL, the integral value of the left bottom region)

In a specific embodiment, the radius of the filtering window is R, and the size of the entire filtering window is (2R+1)*(2R+1). At this time, the bottom running module and the top running module in the window have a distance difference on both sides. For the distance of (2R+1), the integrated values of the left top region TL, the right top region TR, the left bottom region BL, and the right bottom region BR are introduced into the above formula (1) to obtain the entire bilateral filter (2R+1)* ( 2R+1) The sum of the dark primary color image data in the window, and the obtained sum of the dark primary color image data in the window is bilaterally filtered to realize the noise reduction processing of the Bayer image data.

104: Calculate and process the filtered dark primary color image data to obtain Bayer image data with contrast enhancement.

Specifically, first, the histogram statistics are performed according to the RGB image data and the filtered dark primary color image, and the R/G/B image data cumulative histogram, the dark primary color image data histogram, and the dark primary color image data cumulative total value are obtained; /G/B image data accumulation histogram, dark primary image data histogram, dark primary image data cumulative total value for comprehensive analysis, get the gain value and the atmospheric light value corresponding to the R/G/B image data; finally use the Bayer image data The obtained gain value and the atmospheric light value corresponding to the R/G/B image data, the filtered dark primary color image, and the Bayer image data are subjected to comprehensive calculation processing to obtain Bayer image data with contrast enhancement.

Different from the prior art, the bottom, middle and top running modules respectively obtain dark primary color image data of the bottom, middle and top, and then integrate the bottom and top regions to obtain the sum of the dark primary color image data in the filtering window, and comprehensively calculate Processing to obtain Bayer image data with improved contrast. In the above manner, while ensuring the contrast and detail information of the image, the intermediate processing process is simplified, the processing delay is reduced, and the processing process of the Bayer image is improved.

Referring to FIG. 5, FIG. 5 is a schematic structural diagram of an apparatus for improving Bayer image contrast according to a first embodiment of the present invention, which mainly includes: a storage manager 1 for buffering Bayer image line data in a bilateral filtering window, and the storage manager 1 follows The transmission direction includes: a storage controller 18, an arbitration manager 19, and a bottom operation module (not shown) for acquiring bottom dark primary image data of the pixel unit in the Bayer image data, and according to the bottom dark primary color image. The data is calculated to obtain the integrated values of the left bottom region BL and the right bottom region BR, and the bottom running module includes, in order of the transmission direction, a row buffer management module 2, a color filter array interpolation module 3, a dark primary color calculation module 4, and an integral row management module 5; a top running module (not shown) for acquiring top dark primary color image data of the pixel unit in the Bayer image data, and calculating an integral value of the left top region TL and the right top region TR according to the top dark primary color image data The top running module has the same structure as the bottom running module, and the top running module includes, in order of the transmission direction, a line buffer. The management module 14, the color filter array interpolation module 15, the dark primary color calculation module 16 and the integral row management module 17; the middle operation module (not shown) is used to obtain the dark corresponding to the Bayer image row corresponding to the middle portion of the filter window. Primary color image data; at the same time, the Bayer image line and the dark primary color image line are aligned; the middle running module is configured to receive the left bottom area BL integral value, the right bottom area BR integral value, the left top area TL integral value, and the right top The regional TR integral value is integrated into four values of BL, BR, TL, and TR, and the sum of the dark primary color image data in the filtering window is obtained, and then the filtered dark primary color image data is obtained after the bilateral filtering calculation processing; the middle running module includes: The buffer management module 6, the color filter array interpolation module 7, the dark primary color calculation module 8, the barrel shifter 9, the bilateral filter calculation module 10, and the statistical analysis calculation module. The statistical analysis calculation module includes a statistics module 11, an analysis module 12, and a calculation module 13. The bottom running module, the top running module and the middle running module are parallel modules, and both adopt a pipeline structure.

In this embodiment, the line buffer management module (2, 6, and 14 in the figure) is used to rearrange Bayer image data before interpolation; the color filter array interpolation module (3, 7, and 15 in the figure) is used for Bayer image data interpolation calculation. Obtain an RGB image; the dark primary color calculation module (4, 8, 16 in the figure) selects the minimum value in the RGB image as the dark primary image data; the integral row management module (5, 17 in the figure) is used to implement the top of the filtering window. Management of the integration line corresponding to the bottom, and using 2 SRAMs to update and store the integrated value of the simulated dark primary image line data; the barrel shifter 9 is used to specify the direction and displacement of the Bayer image data transmission; bilateral filtering The calculation module 10 is configured to implement filtering and smoothing processing on the dark primary color image data to obtain a filtered dark primary color image; the statistical module 11 is configured to perform histogram statistics on the filtered dark primary color image to obtain R/G/B image data. Cumulative histogram, dark primary image data histogram, dark primary image data cumulative total value; analysis module 12 accumulates R/G/B image data histogram, dark primary image data histogram, dark primary image number According to the cumulative total value, comprehensive analysis is performed to obtain the gain value and the atmospheric light value corresponding to the R/G/B image data; the gain value obtained by the calculation module 13 according to the Bayer image data and the atmospheric light value corresponding to the R/G/B image data and The filtered dark primary color image and the Bayer image data are integrated to obtain Bayer image data with contrast enhancement; the storage controller 18 is used for buffering Bayer image line data in the bilateral filtering window; the arbitration manager 19 is responsible for multiple directions. Bayer image line data read and write management.

In a specific embodiment, the Bayer image data is input to the bottom running module and the storage manager 1, respectively, and the image data is time-divisionally sent to the top running module and the middle running module through the storage controller 18 and the arbitration manager 19; Bayer image data The dark primary image data of the bottom, middle and top of the pixel unit passes through the line buffer management module (2, 6, 14 in the figure), the color filter array interpolation module (3, 7, 15 in the figure) and the dark primary color calculation module ( After processing 4, 8, and 16), the respective dark primary color image data and RGB image data are obtained; the top and bottom dark primary color image data are implemented in the filtering window via the integral row management module (Fig. 5, 17). And the management of the integral line corresponding to the bottom, the integral values of the left top region TL, the right top region TR, the left bottom region BL, and the right bottom region BR are obtained, and an integral map is formed; through the integral map, the left top region TL, the right top The integral values of the region TR, the left bottom region BL and the right bottom region BR are integrated and processed to obtain the sum of the dark primary color image data in the filtering window, and the dark primary color image in the filtering window is obtained. After the data summation is bilaterally filtered by the bilateral filtering calculation module 10, the filtered dark primary color image data is obtained; and the statistic module 11 performs histogram statistics based on the RGB image data and the filtered dark primary color image to obtain an R/G/B image. Data accumulation histogram, dark primary image data histogram, dark primary color image data cumulative total value; R/G/B image data accumulation histogram, dark primary color image data histogram, dark primary color image data cumulative total value by analysis module 12 Performing comprehensive analysis to obtain the gain value and the atmospheric light value corresponding to the R/G/B image data; the gain value obtained by using the Bayer image data in the calculation module 13 and the atmospheric light value corresponding to the R/G/B image data and the filtered The dark primary color image, the Bayer image data, and the comprehensive calculation process obtain the Bayer image data with the contrast enhancement.

In an embodiment, please refer to FIG. 6. FIG. 6 is a schematic structural diagram of an apparatus for improving Bayer image contrast according to a second embodiment of the present invention. The input image data format is an RGB image, and color interpolation array interpolation processing is not required (CFAI, Color). Filter Array Interpolation For the processing, the color filter array interpolation module in the above embodiment can be used to directly obtain the RGB image data for subsequent processing. The specific device structure is similar to the above embodiment, and will not be described here.

In another specific embodiment, please refer to FIG. 7. FIG. 7 is a schematic structural diagram of an apparatus for improving Bayer image contrast according to a third embodiment of the present invention. The input data image is a YUV image, and the YUV format needs to be converted to the RGB format when inputting. The output port of the device 22 needs to be added to the RGB format to the YUV format converter 23. The specific steps are similar to those of the above embodiment, and will not be described here.

The implementation of the Bayer image contrast enhancement device can also be extended to RGB FHD The real-time image processing of the resolution such as (1920*1080) or UD (3840*2160) is not limited herein.

Different from the prior art, the present invention provides a method and a device for improving the contrast of a Bayer image. The method and the device of the present invention adopt a pipeline and data parallel processing structure to simplify the intermediate processing process and reduce the processing while ensuring the contrast and detail information of the image. The delay has improved the processing of Bayer images. Further, the method and device of the present invention converts the integral map required for bilateral filtering into an integral line mode for management, optimizes the dark state image, suppresses the over-bright area, and improves the contrast of the entire image.

The above is only the embodiment of the present invention, and is not intended to limit the scope of the invention, and the equivalent structure or equivalent process transformations made by the description of the invention and the drawings are directly or indirectly applied to other related technologies. The fields are all included in the scope of patent protection of the present invention.

Claims (16)

1. A method for improving the contrast of a Bayer image, including:

   Interpolating current Bayer image data to obtain interpolated RGB image data;

   Obtaining dark primary color image data of the bottom, middle, and top of the pixel unit according to the interpolated RGB image data;

   The acquired dark primary color image data is time-divisionally managed, and the integrated values of the left top region TL and the right top region TR are obtained by the dark primary color image data of the top, and the left bottom region BL and the right are obtained by the dark primary color image data of the bottom. The integral value of the bottom region BR and form an integral map;

   The integrated values of the integral map, the left top region TL, the right top region TR, the left bottom region BL, and the right bottom region BR are integrated to obtain a sum of dark primary color image data in the filtered window. Performing bilateral filtering processing on the sum of the dark primary color image data in the filtering window to obtain filtered dark primary color image data;

   Performing histogram statistics on the RGB image data and the filtered dark primary color image to obtain an R/G/B image data cumulative histogram, a dark primary color image data histogram, and a dark primary color image data cumulative total value;

   Integrating the R/G/B image data accumulation histogram, the dark primary color image data histogram, and the dark primary color image data cumulative total value to obtain a gain value and an atmosphere corresponding to the R/G/B image data. Light value

   The gain value obtained by the Bayer image data and the atmospheric light value corresponding to the R/G/B image data, the filtered dark primary color image, and the Bayer image data are integrated to obtain Bayer image data with contrast enhancement.
2. The method of improving Bayer image contrast according to claim 1, wherein said integral map is implemented by means of integrating lines.
3. The method of improving Bayer image contrast according to claim 1, wherein said passing through said integration map, said left top region TL, said right top region TR, said left bottom region BL, and said right bottom region The step of obtaining the sum of the dark primary color image data in the filtering window after the integral value of the BR is comprehensively processed includes:

   First, the integrated values of the left top region TL and the right bottom region BR are merged, and the integrated values of the right top region TR and the left bottom region BL are subtracted to obtain a sum of dark primary image data in the filtered window.
4. A method for improving the contrast of a Bayer image, including:

   Obtaining dark primary color image data of the bottom, middle, and top of the pixel unit in the current Bayer image data, respectively;

   The acquired dark primary color image data is time-divisionally managed, and the integrated values of the left top region TL and the right top region TR are obtained by the dark primary color image data of the top, and the left bottom region BL and the right are obtained by the dark primary color image data of the bottom. The integral value of the bottom region BR and form an integral map;

   The integrated values of the integral map, the left top region TL, the right top region TR, the left bottom region BL, and the right bottom region BR are integrated to obtain a sum of dark primary color image data in the filtered window. Performing bilateral filtering processing on the sum of the dark primary color image data in the filtering window to obtain filtered dark primary color image data;

   The filtered dark primary color image data is calculated and processed to obtain Bayer image data with improved contrast.
5. The method of improving Bayer image contrast according to claim 4, wherein said integral map is implemented by means of integral lines.
6. The method of improving Bayer image contrast according to claim 4, wherein said passing through said integration map, said left top region TL, said right top region TR, said left bottom region BL, and said right bottom region The step of obtaining the sum of the dark primary color image data in the filtering window after the integral value of the BR is comprehensively processed includes:

   First, the integrated values of the left top region TL and the right bottom region BR are merged, and the integrated values of the right top region TR and the left bottom region BL are subtracted to obtain a sum of dark primary image data in the filtered window.
7. The method for improving the contrast of a Bayer image according to claim 4, wherein the specific steps of respectively obtaining the dark primary color image data of the bottom, middle and top of the pixel unit in the current Bayer image data comprise:

   Interpolating current Bayer image data to obtain interpolated RGB image data;

   According to the interpolated RGB image data, dark primary color image data of the bottom, middle, and top of the pixel unit are respectively acquired.
8. The method for improving the Bayer image contrast according to claim 4, wherein the step of calculating the filtered dark primary color image data to obtain the contrast-enhanced Bayer image data comprises:

   Performing histogram statistics on the RGB image data and the filtered dark primary color image to obtain an R/G/B image data cumulative histogram, a dark primary color image data histogram, and a dark primary color image data cumulative total value;

   Integrating the R/G/B image data accumulation histogram, the dark primary color image data histogram, and the dark primary color image data cumulative total value to obtain a gain value and an atmosphere corresponding to the R/G/B image data. Light value

   The gain value obtained by the Bayer image data and the atmospheric light value corresponding to the R/G/B image data, the filtered dark primary color image, and the Bayer image data are integrated to obtain Bayer image data with contrast enhancement.
9. A device for improving the contrast of a Bayer image, which includes:

   a storage manager, configured to implement buffering of Bayer image line data in a bilateral filtering window;

   a bottom running module, the bottom running module is configured to acquire bottom dark primary color image data of the pixel unit in the Bayer image data, and calculate an integrated value of the left bottom region BL and the right bottom region BR according to the bottom dark primary color image data;

   a top running module, configured to acquire top dark primary color image data of the pixel unit in the Bayer image data, and calculate an integrated value of the left top region TL and the right top region TR according to the top dark primary color image data;

   a middle running module, the middle running module is configured to form an integral map, and the integral of the integral map, the left top region TL, the right top region TR, the left bottom region BL, and the right bottom region BR After the values are comprehensively processed, the sum of the dark primary color image data in the filtering window is obtained, and the sum of the dark primary color image data in the filtering window is bilaterally processed to obtain the filtered dark primary color image data, and the filtered dark primary color image data is used. With calculation processing, Bayer image data with contrast enhancement is obtained.
10. The apparatus for improving the contrast of a Bayer image according to claim 9, wherein: said bottom running module comprises:

    Line buffer management module; Bayer image data color filter array interpolation module for rearranging interpolation;

    a dark primary color calculation module, which selects dark primary color image data by selecting a minimum value in the RGB image;

    a line management module for implementing management of the integral line corresponding to the top and bottom of the filtering window;

    The top running module has the same structure as the bottom running module.
11. The apparatus for improving the contrast of a Bayer image according to claim 9, wherein: said central running module comprises:

    a line buffer management module, a Bayer image data color filter array interpolation module for rearranging interpolation;

    Color filter array interpolation module for Bayer image data interpolation calculation to obtain RGB images;

    a dark primary color calculation module, which selects dark primary color image data by selecting a minimum value in the RGB image;

    a barrel shifter for specifying the direction and displacement of Bayer image data transmission;

    The bilateral filtering calculation module is configured to implement filtering and smoothing processing on the dark primary color image data to obtain a filtered dark primary color image;

    The statistical analysis calculation module is configured to comprehensively analyze and process the dark primary color image data and the filtered dark primary color image data, and obtain Bayer image data with contrast enhancement after calculation.
12. The apparatus for improving Bayer image contrast according to claim 9, wherein the bottom running module, the top running module, and the middle running module are parallel modules.
13. The apparatus for improving Bayer image contrast according to claim 9, wherein said storage manager comprises:

    a storage controller that caches Bayer image line data in a bilateral filtering window;

    The arbitration manager is responsible for reading and writing management of Bayer image line data in multiple directions.
14. The apparatus for improving contrast of a Bayer image according to claim 11, wherein said bilateral filtering calculation module subtracts said right top region by first combining the integrated values of said left top region TL and said right bottom region BR The integrated value of TR and the left bottom region BL is obtained as a sum of dark primary image data in the filter window.
15. The apparatus for improving Bayer image contrast according to claim 11, wherein the dark primary color calculation module acquires dark primary color image data of the bottom, middle, and top of the pixel unit according to the interpolated RGB image data.
16. The apparatus for improving Bayer image contrast according to claim 10, wherein said integral line management module uses two static memories to update and store the integrated value of the simulated dark primary image line data.