Self-adaptive image enhancement method and device and computer storage medium

Abstract

The invention provides a self-adaptive image enhancement method, which comprises the following steps of: partitioning a current image, counting brightness information of each block, and determining a brightness average value Ymax of a sub-block with the brightest brightness average value, a brightness average value Ymin of a sub-block with the darkest brightness average value and a brightness average value Ymean of the whole image; determining the position of a transformation function curve demarcation point (xb, yb) according to the brightness average value and a preset threshold value; determining a transformation function required by enhancement processing according to the demarcation point; the front part of the transformation function at the demarcation point is a nonlinear curve, and the rear part of the transformation function at the demarcation point is a linear curve; and enhancing the brightness value of the current point by using the transformation function. According to the method, parameters are automatically adapted to enhance the image, and smooth transition is achieved. The whole image is uniform in illumination and clear in details.

Description

Technical Field

[0001] The invention belongs to the field of image and video processing, and in particular relates to an adaptive image enhancement method, device and computer storage medium. Background Art

[0002] During the image acquisition process, the image is affected by many factors such as weather, light, instruments and equipment, resulting in uneven overall illumination, insufficient brightness and loss of details. This not only affects the visual effect of the image, but also greatly reduces the reliability of subsequent processing and analysis.

[0003] How to achieve better image display effects in an environment with ever-changing scenes is a problem that needs to be solved urgently. Summary of the invention

[0004] In view of this, the present invention provides an adaptive image enhancement method, device and computer storage medium to solve the deficiencies of the prior art.

[0005] In order to achieve the above object, the present invention adopts the following technical scheme:

[0006] An embodiment of the present invention provides an adaptive image enhancement method, comprising:

[0007] Divide the current image into blocks and count the brightness information of each block to determine the brightness average value Y of the sub-block with the brightest brightness average value max , the average brightness value Y of the sub-block with the darkest brightness value min And the average brightness Y of the entire image mean ;

[0008] Determine the position of the transformation function curve dividing point (xb, yb) according to the brightness average value and the preset threshold value;

[0009] Determine the transformation function required for the enhancement process according to the demarcation point; the transformation function is a nonlinear function in the front part of the demarcation point and a linear function in the rear part;

[0010] The transformation function is used to perform adaptive enhancement processing on the brightness value of the current point.

[0011] Furthermore, a nonlinear function before the dividing point is used to enhance dark details of the image, and a linear function after the dividing point is used to control local highlight overexposure of the image;

[0012] For low-brightness and low-contrast scenes, it mainly improves the overall dark details;

[0013] For low-brightness, high-contrast scenes, it mainly improves local dark details while controlling local highlight overexposure;

[0014] For scenes with medium brightness and medium contrast, the main focus is to improve the overall contrast;

[0015] For scenes with high brightness and high contrast, the main purpose is to control local highlight overexposure;

[0016] For high-brightness, low-contrast scenes, the main focus is to improve the overall contrast while controlling local highlight overexposure.

[0017] Furthermore, the determination of the dividing point is divided into four categories according to the brightness average value and the preset threshold value:

[0018] There are three preset threshold values, namely, a first preset brightness threshold value thH, a second preset brightness threshold value thL and a third preset brightness threshold value thHH, and thHH>thH>thL;

[0019] If the average brightness of the image is brighter, that is, Y mean >thH, the overall image is brighter, and the cutoff point xb is obtained by weighting the average brightness and the minimum brightness, and the weight is controlled by the contrast;

[0020] If the average brightness of the image is at medium brightness, that is, thL <Y mean ≤thH, the overall image is at a medium brightness level, and the cutoff point xb is obtained by weighting the maximum brightness and the average brightness, and the weight is controlled by the contrast;

[0021] If the average brightness of the image is low, that is, Y mean ≤thL, and satisfy Y max >thHH, the cutoff point xb is obtained from the average brightness;

[0022] If the average brightness of the image is low, that is, Y mean ≤thL, and satisfy Y max ≤thHH,

[0023] The dividing point xb is obtained by weighting the average brightness and the maximum brightness, and the weight is controlled by the contrast.

[0024] Furthermore, the position of the dividing point (xb, yb) of the transformation function curve is determined according to the brightness average value and the preset threshold value as follows:

[0025] If Y mean >thH, the expression of the demarcation point (xb, yb) is as follows:

[0026]

[0027] yb=k*(xb-128)+128, where k>1;

[0028] If thL <Y mean≤thH, the expression of the cutoff point (xb, yb) is as follows:

[0029]

[0030] yb=k*(xb-128)+128, k=1;

[0031] If Y mean ≤thL, and Y max >thHH, the expression of the dividing point (xb, yb) is as follows:

[0032] xb=Y mean

[0033] yb=k*(xb-128)+128,0 <k<1;

[0034] If Y mean ≤thL, and Y max ≤thHH, the expression of the dividing point (xb, yb) is as follows:

[0035]

[0036] yb=k*(xb-128)+128,0 <k<1。

[0037] Furthermore, the expression of the transformation function is as follows:

[0038] When x>xb,

[0039]

[0040] When x≤xb,

[0041]

[0042] where k max is the default value, in g is the default value; k max The maximum number of curves with different changing trends is controlled; the g value can be set according to human needs. When the brightness of the low-brightness area needs to be adjusted to be brighter, the larger the g value is set, and the min() function calculates the smaller value of the two.

[0043] Furthermore, the brightness value of the current point is enhanced by:

[0044] First calculate the brightness of each point, according to the brightness value Y of the current point in Substitute the transformation function y = f(x) to calculate the transformed value Y out =f(Y in );

[0045] Then calculate the RGB output value of each point using the following formula:

[0046] R out =R in *Y out / Y in ;

[0047] G out =G in *Y out / Y in ;

[0048] B out =B in *Y out / Y in ; where R in , G in , B in To enhance the RGB value before processing, R out , G out , B out It is the RGB value after enhancement.

[0049] Furthermore, when enhancing the video stream image, the calculation of the transformation function should take into account not only the image brightness information of the current frame, but also the image brightness information of the previous frame; it is necessary to perform inter-frame filtering on the dividing point (xb, yb) of the transformation function curve; specifically, the following formula is used:

[0050] xb(i)=(xb(i-1)+xbc) / 2

[0051] yb(i)=(yb(i-1)+ybc) / 2

[0052] Among them, xbc, ybc are calculated by the brightness statistics of the current frame, xb(i-1), yb(i-1) are the dividing points used in the previous frame, and xb(i), yb(i) are the dividing points finally used after inter-frame filtering at the current moment.

[0053] Furthermore, the key value in the method also takes into account the image brightness information of the previous frame; it is necessary to perform inter-frame filtering on the key value of the nonlinear curve segment of the transformation function curve; after the Key value is calculated, the key value is inter-frame filtered in the following manner: Specifically, the following formula is used:

[0054] key(i)=(key(i-1)+keyc) / 2

[0055] Among them, keyc is calculated by the brightness statistics of the current frame, key(i-1) is the key value used in the previous frame, and key(i) is the key value finally used after inter-frame filtering at the current moment.

[0056] The embodiment of the present invention further provides an adaptive image enhancement device, comprising:

[0057] Brightness statistics module: used to divide the current image into blocks and count the brightness information of each block, and determine the brightness average value Y of the sub-block with the brightest brightness average value max , the average brightness value Y of the sub-block with the darkest brightness value min And the average brightness Y of the entire image mean ;

[0058] A demarcation point determination module: used to determine the position of the demarcation point (xb, yb) of the transformation function curve according to the brightness average value and the preset threshold value;

[0059] Transformation function calculation module: used to determine the transformation function required for enhancement processing according to the demarcation point; the transformation function is a nonlinear function in the front part of the demarcation point, and a linear function in the rear part;

[0060] Pixel enhancement processing module: used to use the transformation function to perform adaptive enhancement processing on the brightness value of the current point.

[0061] Furthermore, the demarcation point determination module determines the demarcation point in the following manner:

[0062] If Y mean >thH, the expression of the demarcation point (xb, yb) is as follows:

[0063]

[0064] yb=k*(xb-128)+128, where k>1

[0065] If thL <Y mean ≤thH, the expression of the cutoff point (xb, yb) is as follows:

[0066]

[0067] yb=k*(xb-128)+128,k=1

[0068] If Y mean ≤thL, and Y max >thHH, the expression of the dividing point (xb, yb) is as follows:

[0069] xb=Y mean

[0070] yb=k*(xb-128)+128,0 <k<1

[0071] If Y mean ≤thL, and Y max ≤thHH, the expression of the dividing point (xb, yb) is as follows:

[0072]

[0073] yb=k*(xb-128)+128,0 <k<1;

[0074] Wherein thH, thL and thHH are respectively a first preset brightness threshold value, a second preset brightness threshold value and a third preset brightness threshold value, and thHH>thH>thL.

[0075] Furthermore, the transformation function calculation module determines the transformation function in the following manner:

[0076] When x>xb,

[0077]

[0078] When x≤xb,

[0079]

[0080] where k max is the default value, in g is the default value; k max The maximum number of curves with different changing trends is controlled; the g value can be set according to human needs. When the brightness of the low-brightness area needs to be adjusted brighter, the larger the g value is set.

[0081] An embodiment of the present invention further provides a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned adaptive image enhancement method is implemented.

[0082] The adaptive image enhancement method provided by the present invention aims to adaptively enhance the brightness and contrast of an image and improve the overall clarity of the image, especially for video images with changing scenes, which can automatically adapt parameters to enhance the image and make a smooth transition. The overall illumination of the image is uniform and the details are clear. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0084] Figure 1 A schematic diagram of a flow chart of an adaptive image enhancement method provided by an embodiment of the present invention;

[0085] Figure 2 is a schematic diagram of the transformation function in this embodiment;

[0086] Figure 3 A schematic diagram of the structure of an adaptive image enhancement device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0087] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0088] Figure 1 The embodiment of the present invention provides an adaptive image enhancement method, comprising:

[0089] S1. Divide the current image into blocks and count the brightness information of each block to determine the brightness average value Y of the sub-block with the brightest brightness average value max , the average brightness value Y of the sub-block with the darkest brightness value min And the average brightness Y of the entire image mean .

[0090] In this embodiment, a single frame image is taken as an example, and the brightness information at the current moment is calculated from the single frame image. Specifically, the input image is divided into M*N image sub-blocks. In this specific embodiment, the image can be divided into 8*8 sub-blocks. For each image sub-block, the brightness average of all pixels in it is calculated. The sub-block with the brightest brightness average is found, and the brightness average of the sub-block is recorded as Y max Find the sub-block with the darkest average brightness, and record the average brightness of this sub-block as Y min ; Calculate the average brightness of the entire image, recorded as Y mean .

[0091] S2. Determine the position of the demarcation point (xb, yb) of the transformation function curve based on the average brightness value and a preset threshold value. The preset threshold value is an engineering experience value, and there are three of them, namely the first preset brightness threshold value thH, the second preset brightness threshold value thL, and the third preset brightness threshold value thHH, and thHH > thH > thL.

[0092] The enhancement curve of the transformation function is mainly divided into two segments. The segment before the demarcation point is used to enhance the details in the dark area, and the segment after the demarcation point is used to control the overexposure of local highlights.

[0093] For scenes with low brightness and low contrast, mainly enhance the overall details in the dark area; for scenes with low brightness and high contrast, mainly enhance the local details in the dark area and control the overexposure of local highlights at the same time; for scenes with medium brightness and medium contrast, mainly enhance the overall contrast; for scenes with high brightness and high contrast, mainly control the overexposure of local highlights; for scenes with high brightness and low contrast, mainly enhance the overall contrast and control the overexposure of local highlights at the same time.

[0094] As Figure 2 shown, first determine the position of the demarcation point (xb, yb). The demarcation point divides the curve into two parts. The first part is a non-linear curve, mainly used to control the adjustment of the contrast in the low-brightness area to enhance the details in the dark area. The second part after the demarcation point is a linear curve, mainly used to control the contrast adjustment in the high-brightness area to control the overexposure of local highlights. And the position of the demarcation point determines the overall adjustment focus.

[0095] When the demarcation point xb < yb, the contrast of the first part before the demarcation point increases, and the contrast of the second part after the demarcation point decreases; when the demarcation point xb > yb, the contrast of the low-brightness interval in the first part before the demarcation point increases, the contrast near the demarcation point decreases, and the contrast of the second part after the demarcation point increases. When the demarcation point xb = yb, the contrast of the low-brightness interval in the first part before the demarcation point increases, the contrast of the medium-brightness interval decreases, and the contrast of the second part after the demarcation point remains unchanged. Specifically, determine the position of the demarcation point according to the average brightness and contrast of the image. The average brightness of the image can be represented by Y mean characterization. The larger the value of Y mean , the greater the overall average brightness of the image. The smaller the value of Y mean , the smaller the overall average brightness of the image. The contrast can be represented by Y max -Y min . The larger the value of Y max -Y min , the greater the corresponding contrast. The smaller the value of Y max -Y min , the smaller the corresponding contrast. The average brightness of the image can be represented by Y mean characterization.

[0096] Among them, if the average brightness is brighter, that is, Y mean >thH, the overall image is brighter, and the boundary point xb is determined by the average brightness Y mean and minimum brightness Y min The weight is controlled by contrast, and the expression of the dividing point is as follows:

[0097]

[0098] yb=k*(xb-128)+128,k>1

[0099] If the average brightness is at medium brightness, that is, thL <Y mean ≤thH, the overall image is at a medium brightness level, and the boundary point xb is the maximum brightness Y max and the average brightness Y mean The weight is controlled by contrast, and the expression of the dividing point is as follows:

[0100]

[0101] yb=k*(xb-128)+128,k=1

[0102] If the average brightness is at a low brightness, that is, Y mean ≤thL, and satisfy Y max >thHH, the demarcation point xb is obtained by the average brightness, and the expression of the demarcation point is as follows:

[0103] xb=Y mean

[0104] yb=k*(xb-128)+128,0 <k<1

[0105] If the average brightness is at a low brightness, that is, Y mean ≤thL, and satisfy Y max ≤thHH,

[0106] The dividing point xb is determined by the average brightness Y mean and maximum brightness Y max The weight is controlled by contrast, and the expression of the dividing point is as follows:

[0107]

[0108] yb=k*(xb-128)+128,0 <k<1。

[0109] S3. Determine the transformation function required for enhancement processing according to the demarcation point.

[0110] The transformation function is a nonlinear curve in the front part of the demarcation point, and a linear curve in the rear part. Figure 2 As shown in the figure, once the dividing point is determined, the curve of the linear transformation of the latter part can be determined. Specifically, when x>xb:

[0111]

[0112] When x≤xb:

[0113]

[0114] in Among them, in this embodiment, k max =512,k max The maximum number of curves with different changing trends is controlled. In this embodiment, g=256, and the g value can be set according to human needs. When the brightness of the low-brightness area needs to be adjusted to be brighter, the g value is set to be larger. The min() function operates to take the smaller value of the two. Specifically, when the contrast is large, the key value is large, and the curve changes steeply. Conversely, when the contrast is small, the key value is small, and the curve changes slowly. Similarly, when the overall brightness of the image is low, the key value is large, and the curve changes steeply. Conversely, when the overall brightness of the image is high, the key value is small, and the curve changes slowly. Figure 2 As shown, the blue curve and the red curve have different changing trends.

[0115] S4. Use the transformation function to perform adaptive enhancement processing on the brightness value of the current point.

[0116] First, calculate the brightness of each point, according to the brightness value Y of the current point in and the current transformation function (the function given in step S3) calculates the transformed value Y out Then calculate the RGB output value of each point, specifically, using the following formula:

[0117] R out =R in *Y out / Y in ;

[0118] G out =G in *Y out / Y in ;

[0119] B out =B in *Y out / Y in ; where R in , G in , B in To enhance the RGB value before processing, Rout , G out , B out It is the RGB value after enhancement.

[0120] When enhancing the video stream image, the calculation of the transformation function must take into account not only the image brightness information of the current frame, but also the image brightness information of the previous frame. In order to ensure the smooth adjustment of the video stream image, the dividing point and key value of the curve are filtered between frames. Specifically, as shown in the following formula:

[0121] xb(i)=(xb(i-1)+xbc) / 2

[0122] yb(i)=(yb(i-1)+ybc) / 2

[0123] key(i)=(key(i-1)+keyc) / 2

[0124] Among them, xbc, ybc, and keyc are calculated by the image brightness statistics of the current frame, that is, xbc and ybc are solved according to the formula in step S2, and the formula in step S3 is used to calculate Solve keyc; xb(i-1), yb(i-1), key(i-1) are the demarcation point and key value finally adopted after inter-frame filtering of the previous frame; xb(i), yb(i), key(i) are the demarcation point and key value finally adopted at the current moment.

[0125] This method aims to adaptively enhance the brightness and contrast of the image and improve the overall clarity of the image. Especially for video images with changing scenes, the parameters can be automatically adapted to enhance the image and achieve smooth transition.

[0126] like Figure 3 As shown, an embodiment of the present invention further provides an adaptive image enhancement device, including: a brightness statistics module, a demarcation point determination module, a transformation function calculation module and a pixel enhancement processing module.

[0127] The brightness statistics module is used to divide the current image into blocks and count the brightness information of each block, and determine the brightness average value Y of the sub-block with the brightest brightness average value. max , the average brightness value Y of the sub-block with the darkest brightness value min And the average brightness Y of the entire image mean .

[0128] The demarcation point determination module is used to determine the position of the demarcation point (xb, yb) of the transformation function curve according to the brightness average value and the preset threshold value.

[0129] The transformation function calculation module is used to determine the transformation function required for enhancement processing according to the demarcation point; the transformation function is a nonlinear curve in the front part of the demarcation point, and a linear curve in the rear part.

[0130] The pixel enhancement processing module is used to use the transformation function to perform adaptive enhancement processing on the brightness value of the current point.

[0131] It should be noted that: the adaptive image enhancement device provided in the above embodiment only uses the division of the above program modules as an example to illustrate when performing image enhancement. In actual applications, the above processing can be assigned to different program modules as needed, that is, the internal structure of the device is divided into different program modules to complete all or part of the processing described above. In addition, the adaptive image enhancement device provided in the above embodiment and the adaptive image enhancement method embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment. The beneficial effects are the same as those in the method embodiment, and will not be repeated here.

[0132] The embodiment of the present invention further provides a computer storage medium, which is a computer-readable storage medium, on which a computer program is stored, and the computer program can be executed by a processor of an adaptive image enhancement device to complete the steps of the adaptive image enhancement method. The computer-readable storage medium can be a magnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM).

[0133] In the several embodiments provided by the present invention, it should be understood that the disclosed methods and intelligent devices can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0134] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An adaptive image enhancement method, characterized in that: include: Divide the current image into blocks and count the brightness information of each block to determine the brightness average value Y of the sub-block with the brightest brightness average value max , the average brightness value Y of the sub-block with the darkest brightness value min And the average brightness Y of the entire image mean ; Determine the position of the transformation function curve dividing point (xb, yb) according to the brightness average value and the preset threshold value; Determine the transformation function required for the enhancement process according to the demarcation point; the transformation function is a nonlinear function in the front part of the demarcation point and a linear function in the rear part; The transformation function is used to perform adaptive enhancement processing on the brightness value of the current point.

2. The method according to claim 1, characterized in that: The nonlinear function before the dividing point is used to enhance the dark details of the image, and the linear function after the dividing point is used to control the local highlight overexposure of the image; For low-brightness and low-contrast scenes, adaptive image enhancement mainly improves overall dark details; For low-brightness, high-contrast scenes, adaptive image enhancement mainly improves local dark details while controlling local highlight overexposure; For scenes with medium brightness and medium contrast, adaptive image enhancement mainly improves the overall contrast; For scenes with high brightness and high contrast, adaptive image enhancement mainly controls local highlight overexposure; For high-brightness, low-contrast scenes, adaptive image enhancement mainly improves the overall contrast while controlling local highlight overexposure.

3. The method according to claim 1, characterized in that: The determination of the dividing point is divided into four categories according to the brightness average value and the preset threshold value: There are three preset threshold values, namely, a first preset brightness threshold value thH, a second preset brightness threshold value thL and a third preset brightness threshold value thHH, and thHH>thH>thL; If the average brightness of the image is brighter, that is, Y mean >thH, the overall image is brighter, and the cutoff point xb is obtained by weighting the average brightness and the minimum brightness, and the weight is controlled by the contrast; If the average brightness of the image is at medium brightness, that is, thL <Y mean ≤thH, the overall image is at a medium brightness level, and the cutoff point xb is obtained by weighting the maximum brightness and the average brightness, and the weight is controlled by the contrast; If the average brightness of the image is low, that is, Y mean ≤thL, and satisfy Y max >thHH, the cutoff point xb is obtained from the average brightness; If the average brightness of the image is low, that is, Y mean ≤thL, and satisfy Y max ≤thHH, the dividing point xb is obtained by weighting the average brightness and the maximum brightness, and the weight is controlled by the contrast.

4. The method according to claim 3, characterized in that The position of the transformation function curve demarcation point (xb, yb) is determined according to the brightness average value and the preset threshold value as follows: If Y mean >thH, the expression of the demarcation point (xb, yb) is as follows: yb=k*(xb-128)+128, where k>1; If thL <Y mean ≤thH, the expression of the cutoff point (xb, yb) is as follows: yb=k*(xb-128)+128, k=1; If Y mean ≤thL, and Y max >thHH, the expression of the dividing point (xb, yb) is as follows: xb=Y mean yb=k*(xb-128)+128,0 <k<1; If Y mean ≤thL, and Y max ≤thHH, the expression of the dividing point (xb, yb) is as follows: yb=k*(xb-128)+128,0 <k<1。 5. The method according to claim 1, characterized in that The expression of the transformation function is as follows: When x>xb, When x≤xb, where k max is the default value, in g is the default value; k max The maximum number of curves with different changing trends is controlled; the g value can be set according to human needs. When the brightness of the low-brightness area needs to be adjusted to be brighter, the larger the g value is set, and the min() function calculates the smaller value of the two.

6. The method according to claim 1, characterized in that The enhancement process of the brightness value of the current point specifically includes: First calculate the brightness of each point, according to the brightness value Y of the current point in Substitute the transformation function y = f(x) to calculate the transformed value Y out =f(Y in ); Then calculate the RGB output value of each point using the following formula: R out =R in *Y out / Y in ; G out =G in *Y out / Y in ; B out =B in *Y out / Y in ; where R in , G in , B in To enhance the RGB value before processing, R out , G out , B out It is the RGB value after enhancement.

7. The method according to claim 4, characterized in that When enhancing the video stream image, the calculation of the transformation function should take into account not only the image brightness information of the current frame, but also the image brightness information of the previous frame; it is necessary to perform inter-frame filtering on the dividing point (xb, yb) of the transformation function curve; specifically, the following formula is used: xb(i)=(xb(i-1)+xbc) / 2 yb(i)=(yb(i-1)+ybc) / 2 Among them, xbc, ybc are calculated by the brightness statistics of the current frame, xb(i-1), yb(i-1) are the dividing points used in the previous frame, and xb(i), yb(i) are the dividing points finally used after inter-frame filtering at the current moment.

8. The method according to claim 5, characterized in that In the method, the key value also takes into account the image brightness information of the previous frame; it is necessary to perform inter-frame filtering on the key value of the nonlinear curve segment of the transformation function curve; after the Key value is calculated, the key value is subjected to inter-frame filtering in the following manner: Specifically, the following formula is used: key(i)=(key(i-1)+keyc) / 2 Among them, keyc is calculated by the brightness statistics of the current frame, key(i-1) is the key value used in the previous frame, and key(i) is the key value finally used after inter-frame filtering at the current moment.

9. An adaptive image enhancement device, characterized in that: include: Brightness statistics module: used to divide the current image into blocks and count the brightness information of each block, and determine the brightness average value Y of the sub-block with the brightest brightness average value max , the average brightness value Y of the sub-block with the darkest brightness value min And the average brightness Y of the entire image mean ; A demarcation point determination module: used to determine the position of the demarcation point (xb, yb) of the transformation function curve according to the brightness average value and the preset threshold value; Transformation function calculation module: used to determine the transformation function required for enhancement processing according to the demarcation point; the transformation function is a nonlinear function in the front part of the demarcation point, and a linear function in the rear part; Pixel enhancement processing module: used to use the transformation function to perform adaptive enhancement processing on the brightness value of the current point.

10. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the adaptive image enhancement method according to any one of claims 1 to 8 is implemented.

Images