Glare suppression low-illumination image enhancement method and system based on global mapping and local tuning

Abstract

The invention also provides a glare suppression low-illumination image enhancement method based on global mapping and local tuning, and the method comprises the following steps: S1, carrying out the preliminary enhancement of an original image, and adjusting the color saturation and contrast of the image; s2, carrying out image region gradient detection on the original image to calculate a region with relatively violent brightness change, and adaptively carrying out further enhancement on the image through a corresponding algorithm; and S3, superposing the images before and after enhancement through a linear expansion algorithm and a pixel-level weight adaptive algorithm so as to suppress glare in the images, and adaptively adjusting the contrast and color in the images after glare suppression by adopting the algorithm so as to enable the images to accord with the visual habits of human eyes. The technical scheme has the advantages of being fast, good in robustness and suitable for various dark and glare scenes, and good enhancement effect and real-time performance can be guaranteed under various low-illumination scenes and image resolutions.

Description

Technical Field

[0001] The present invention relates to the technical field of image enhancement, and in particular to a glare suppression low-illuminance image enhancement method and system based on global mapping and local optimization. Background Art

[0002] With the rapid development of the information age, it has become crucial to obtain information effectively and quickly. As an important information medium, images play an important role in people's daily lives. Computer-based high-level visual applications have high requirements for image quality, such as security monitoring, face recognition, target detection, and intelligent driving. However, due to inevitable environmental factors, such as undesirable lighting conditions at night and backlight, or due to equipment defects and limitations of shooting technology, the acquired images have many problems such as buried scene content, reduced contrast, severe noise pollution, and inaccurate colors. These characteristics reduce the usability of such images, which not only seriously affects human subjective feelings, but also greatly limits subsequent computer vision tasks. Therefore, it is of great significance to carry out research on low-light image enhancement methods and further obtain and utilize information in images.

[0003] Low-light images often have low overall brightness, and the overall outline and detailed texture of the image are difficult to identify, resulting in less information and lower contrast in the image. At the same time, noise can easily cover up the detailed texture of the image, so it is difficult for us to directly obtain good quality image data from low-light images.

[0004] Existing low-light image enhancement methods can be roughly divided into two categories: traditional methods and deep learning-based methods. Among them, traditional methods are usually based on simple and fixed rules and prior knowledge, so their enhancement effects are often unsatisfactory and image information and details are easily lost; and some methods usually require multiple iterations to obtain better enhancement results, so their real-time performance is poor. For deep learning methods, it is difficult to balance the enhancement effect and model size, so it is difficult to apply in practice. In addition, the existing methods usually do not pay attention to the glare and light spot problems in low-light images. Many methods are often prone to magnifying the light spots and halos in the picture, resulting in a poor visual experience. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a glare suppression low-illuminance image enhancement method and system based on global mapping and local tuning, which has the advantages of being fast, robust, and adaptable to a variety of dark and glare scenes, and can ensure good enhancement effect and real-time performance under a variety of low-illuminance scenes and image resolutions.

[0006] To achieve the above object, the present invention adopts the following technical solution: a glare suppression low-light image enhancement method based on global mapping and local tuning, comprising the following steps:

[0007] Step S1: Perform preliminary enhancement on the original image, perform global tone mapping through an improved algorithm parameter optimization strategy to restore certain details and information of the original image, and adjust the color saturation and contrast of the image;

[0008] Step S2: performing image region gradient detection on the original image to calculate the region with more drastic changes in brightness and darkness, and further enhancing the image adaptively through a corresponding algorithm;

[0009] Step S3: The images before and after enhancement are superimposed by a linear extension algorithm and a pixel-level weight adaptive algorithm, thereby suppressing the glare in the image, and the algorithm is used to adaptively adjust the contrast and color in the image after glare suppression, so that the image conforms to the visual habits of the human eye.

[0010] In a preferred embodiment, the step S1 specifically includes the following steps:

[0011] Step S11: The specific calculation formula for converting the original image from the RGB space to the HSV space is as follows:

[0012] V=max(R,G,B) (1)

[0013]

[0014] Among them, H represents the hue component of the HSV space, which refers to the type or hue of the color, S represents the saturation component of the HSV space, which refers to the vividness of the color, and V represents the brightness component of the HSV space, which refers to the brightness of the color;

[0015] Step S12: Calculate the logarithmic mean value Y of the brightness using the V component calculated in step S11 ave :

[0016]

[0017] Where MN represents the product of the height and width of the original image, σ represents a constant, x represents the horizontal coordinate of the pixel, and y represents the vertical coordinate of the pixel;

[0018] Step S13: Calculate the standard deviation δ of all pixels in the V component using formula (5), where v(i, j) represents a pixel on the V component, and μ represents the pixel mean of the original image;

[0019]

[0020] Step S14: according to different pixel standard deviations, the input original image is classified according to formula (6), and the input original image is classified into three categories: an image with non-uniform illumination, an image with uniform illumination, and an image with brightness difference between the two;

[0021]

[0022] Where δ represents the standard deviation of the input original image pixels, δ1 = 0.17, δ2 = 0.09;

[0023] Step S15: Different L ave Different adjustment functions are used to perform adjustments to enhance the curvature of the global tone mapping function curve, thereby enhancing the adaptability to various dark scenes with different brightness conditions.

[0024] In a preferred embodiment, the step S1 further includes:

[0025] Step S16: The optimization adjustment strategy of the brightness logarithm mean can be specifically represented by the following piecewise function, where Y c Represents the logarithmic mean of brightness after adjustment by the adjustment function;

[0026]

[0027] Step S17: Introduce parameter P1 to adjust Y c and V max , the adjustment formula is as follows:

[0028] Y cln =ln(P1Y c +1)

[0029] V maxln =ln(P1V max +1) (8)

[0030] where Y cln It represents the logarithmic mean of brightness obtained after the above adjustment, V maxln represents the maximum brightness obtained after the above adjustment, P1 represents an empirical parameter used to adjust the representation ability of the enhancement curve to improve the final effect of the present invention, and P1 is set to 13 in the present invention;

[0031] Step S18: Using the obtained Y cln and V maxln Substitute the following formula for calculation and introduce the adjustment parameter P2, where L e Represents the result obtained after preliminary enhancement of global tone mapping;

[0032]

[0033] Step S19: Calculate the enhancement curve cutoff threshold V close to the value obtained by the P1 and P2 parameters using the Newton iteration algorithm. cut , where the loop parameter of the Newton iteration method is set to 20 in the present invention, and only the pixel part smaller than the threshold value will be enhanced by the enhancement curve, so we get:

[0034]

[0035] In a preferred embodiment, the step S2 further includes:

[0036] Step S21: using the Scharr operator to perform a convolution operation on the brightness component V obtained in the global tone mapping module, thereby extracting the gradient information at each pixel and obtaining a gradient weight information map in the x direction and the y direction;

[0037] Step S22: The size of the Scharr operator is 3×3. The difference between pixels is magnified by the weight coefficient in the filter. The Scharr in the x direction and the y direction are defined as follows:

[0038]

[0039] Step S23: Use the two Scharr operators in step S22 to perform convolution operations on the brightness component V, and obtain two matrices g with the same size as the original image. x and g y The specific calculation formula is as follows, where G x () represents the convolution operation on the image, and the convolution kernel is G x , G y () represents the convolution operation on the image, and the convolution kernel is G y ;

[0040] g x (x,y)=G x (V(x,y)),g y (x,y)=G y (V(x,y)) (12);

[0041] Step S24: Obtain an enhanced weight map of the gradient information according to the following formula, where σ represents a very small constant to avoid the situation where the gradient is 0;

[0042]

[0043] Step S25: The final enhancement weight map is obtained by combining the brightness component V and the re weight map obtained in the previous step and performing a power law transformation, where β = 255, which is used to normalize the brightness component V to between (0, 1), and α is an empirical parameter;

[0044]

[0045] Step S26: The gradient detection local enhancement module is used to further enhance the image after global tone mapping, and can restore image detail information. The image after local enhancement is obtained by the following formula:

[0046]

[0047] In a preferred embodiment, the step S3 specifically includes:

[0048] Step S31: using a linear expansion algorithm to suppress glare;

[0049] Step S32: Using a linear expansion algorithm that matches the human eye, as shown in the following formula, where L e It represents the image enhanced by the above two modules, L o represents the original image, a1 and a2 represent the corresponding weights;

[0050] L p =a1L o (x,y)+a2L e (x,y) (16)

[0051] L p Represents the result obtained after the linear expansion algorithm;

[0052] Step S33: The weights a1 and a2 in the linear extension algorithm have different values ​​at different brightness positions. In brighter pixel areas, the weight of the original image is higher, and in darker pixel areas, the weight of the original image is lower; wherein the threshold T = 256-V ave , a2=1-a1, where V ave Indicates the average value of image brightness;

[0053]

[0054] Step S34: using the CLAHE algorithm to further improve the image contrast while retaining the glare suppression effect;

[0055] Step S35: Convert the image from RGB space to XYZ space, then from XYZ space to Lab space, and then apply the CLAHE algorithm to the L component. The formulas used for space conversion are shown in (18), (19), and (20):

[0056]

[0057]

[0058] In a preferred embodiment, the step S3 specifically further includes:

[0059] Step S36: Apply the CLAHE algorithm to the L component obtained in step S35, set the contrast l in the CLAHE algorithm to 1, and set the size of the divided sub-images to 10×10. After the CLAHE algorithm is enhanced, the final output image is obtained, as shown below, where L Le Represents the L component after Le is converted to Lab space;

[0060] L out =CLAHE(L Le ,l,size) (21)

[0061] Step S37: The L component obtained after contrast processing of the image and other unprocessed components are combined and converted back to the RGB space to obtain the final enhanced image of the present invention.

[0062] The present invention also provides a glare suppression low-illuminance image enhancement system based on global mapping and local tuning, which runs the above-mentioned glare suppression low-illuminance image enhancement method based on global mapping and local tuning; it also includes a global tone mapping enhancement module, a gradient detection local enhancement module and a glare suppression module.

[0063] In a preferred embodiment, the global tone mapping enhancement module is responsible for performing preliminary enhancement on the original image, performing global tone mapping through an improved algorithm parameter optimization strategy to restore certain details and information of the original image, and adjusting the color saturation and contrast of the image.

[0064] In a preferred embodiment, the gradient detection local enhancement module performs image region gradient detection on the original image to calculate the region with more drastic changes in brightness and darkness, and further enhances the image adaptively through a corresponding algorithm.

[0065] In a preferred embodiment, the glare suppression module uses a linear expansion algorithm to superimpose the images before and after enhancement through a pixel-level weight adaptive algorithm, thereby suppressing glare in the image, and uses an algorithm to adaptively adjust the contrast and color in the image after glare suppression to make the image conform to the visual habits of the human eye.

[0066] Compared with the prior art, the present invention has the following beneficial effects: Compared with the existing low-light image enhancement method, the present invention combines low-light image enhancement with glare suppression, taking into account various night scenes and lighting conditions, and can not only effectively restore the details of the darker areas in the image, but also will not amplify the bright spots and halos that already exist in the picture, and can suppress the brighter areas in the image to a certain extent, achieving brightness balance for the entire image. In addition, the present invention has the characteristics of high real-time performance while having a good enhancement effect, and balances the enhancement effect and enhancement efficiency, and can guarantee good real-time performance at a variety of image resolution sizes. In summary, the present invention can take into account the restoration tasks of multiple lighting scenes at the same time, has a good enhancement effect, and can guarantee good real-time performance and be applied in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A schematic diagram of a technical route of a preferred embodiment of the present invention;

[0068] Figure 2 A specific flow chart of a preferred embodiment of the present invention;

[0069] Figure 3 A graph showing the relationship between the logarithmic mean of brightness and the enhancement value in a preferred embodiment of the present invention;

[0070] Figure 4 This is a schematic diagram comparing the effects of the present invention and other low-light enhancement methods according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0071] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0072] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0073] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application; as used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or their combinations.

[0074] like Figure 1-4 As shown, the method of the present invention can be specifically divided into the following modules: a global tone mapping enhancement module, a gradient detection local enhancement module, and a glare suppression module.

[0075] Global tone mapping module: This module is responsible for the preliminary enhancement of the original image. It performs global tone mapping through an improved algorithm parameter optimization strategy to restore certain details and information of the original image, and adjusts the color saturation and contrast of the image.

[0076] Gradient weighted local enhancement module: This module performs image region gradient detection on the original image to calculate the area with more drastic changes in brightness and darkness, and further enhances the image adaptively through the corresponding algorithm;

[0077] Glare suppression module: The task of this module is to suppress the glare in the image by superimposing the images before and after enhancement through a linear expansion algorithm and a pixel-level weight adaptive algorithm, and to use an algorithm to adaptively adjust the contrast and color of the image after glare suppression to make the image conform to the visual habits of the human eye;

[0078] Specifically, the specific steps of the global tone mapping module are as follows:

[0079] 1) The three channels of HSV space represent the hue, saturation and brightness of the image respectively. Compared with RGB space, HSV space is more consistent with the human eye's perception of color, and converting the image to HSV space can quickly obtain the image brightness component, which is beneficial to the subsequent calculation of the module. The specific calculation formula for converting the original image from RGB space to HSV space is as follows:

[0080] V=max(R,G,B) (1)

[0081]

[0082] Among them, H represents the hue component of the HSV space, which refers to the type or hue of the color, S represents the saturation component of the HSV space, which refers to the vividness of the color, and V represents the brightness component of the HSV space, which refers to the brightness of the color;

[0083] 2) Use the V component calculated in the previous step to calculate the logarithmic average value Y of the brightness ave , the calculation formula is as follows, where MN represents the product of the height and width of the original image, and σ represents a very small constant to avoid that all pixels of the original image are black pixels with a pixel value of 0;

[0084]

[0085] Where MN represents the product of the height and width of the original image, σ represents a constant, x represents the horizontal coordinate of the pixel, and y represents the vertical coordinate of the pixel;

[0086] 3) The standard deviation of all pixels in a brightness image can reflect the degree of uniform illumination of the image to a certain extent. The standard deviation δ of all pixels in the V component is calculated using the following formula, where v(i,j) represents a pixel on the V component and μ represents the pixel mean of the original image;

[0087]

[0088] 4) According to different pixel standard deviations, the present invention classifies the input original image according to the following formula, and classifies the input original image into three categories: images with non-uniform illumination, images with uniform illumination, and images with brightness differences between the two. Where δ1 = 0.17, δ2 = 0.09;

[0090]

[0091] Where δ represents the standard deviation of the input original image pixels, δ1 = 0.17, δ2 = 0.09;

[0092] 5) The present invention uses the value of the standard deviation δ of the pixel to calculate the different L ave Different adjustment functions are used to enhance the curvature of the global tone mapping function curve, thereby enhancing the adaptability to various dark scenes with different brightness conditions. The final pixel enhancement curve changes are shown in the figure below. The three different curves in the figure represent different L ave The relationship between pixel value and enhancement value under the condition (the blue curve corresponds to L ave Maximum, the orange curve corresponds to L ave Secondly, the green curve corresponds to L ave It can be seen that as the mean value of the logarithmic brightness decreases, the enhancement ratio of the global tone mapping for darker pixels gradually increases, while the enhancement ratio for brighter pixels is very small. Therefore, the mean value of the logarithmic brightness can be adjusted to achieve a better enhancement effect.

[0093] 6) The optimization adjustment strategy of the brightness logarithmic mean can be specifically represented by the following piecewise function, where Y c It represents the logarithmic mean of brightness after adjustment by the adjustment function, where δ1=0.17,

[0094] δ2=0.09;

[0095]

[0096] 7) According to the above formula, we get Y c In order to improve the algorithm's ability to enhance darker images and suppress brighter images, the parameter P1 is introduced to adjust Y c and V max , the adjustment formula is as follows:

[0097]

[0098] where Y cln It represents the logarithmic mean of brightness obtained after the above adjustment, V maxln represents the maximum brightness obtained after the above adjustment, P1 represents an empirical parameter used to adjust the representation ability of the enhancement curve to improve the final effect of the present invention, and P1 is set to 13 in the present invention;

[0099] 8) According to the Weber-Fechner theorem, the human eye's perception of images is more consistent with the characteristics of the log function, so we can use the Y obtained in the previous step cln and V maxln Substitute the following formula for calculation and introduce the adjustment parameter P2, where L e It represents the result obtained by the preliminary enhancement after the global tone mapping is performed by the present invention;

[0100]

[0101] 9) In the present invention, the parameters P1 and P2 are generally set to 10 and 0.85. In addition, based on the previous step, the present invention uses the Newton iteration algorithm to compress the global tone enhancement curve to further improve the algorithm performance. Specifically, the present invention uses the Newton iteration algorithm to calculate the enhancement curve cutoff threshold V close to the P1 and P2 parameters. cut , where the loop parameter of the Newton iteration method is set to 20 in the present invention, and only the pixel part smaller than the threshold value will be enhanced by the enhancement curve, so it can be obtained:

[0102]

[0103] Specifically, the gradient weight local enhancement module:

[0104] 1) A low-light image often has a low contrast, and many valuable details in the image are difficult to recover after global tone mapping, while gradient information can well characterize the brightness changes, edge contours and other information in the image. The present invention uses the Scharr operator to perform a convolution operation on the brightness component V obtained in the global tone mapping module, thereby extracting the gradient information at each pixel and obtaining the gradient weight information map in the x-direction and y-direction.

[0105] 2) The size of the Scharr operator is 3×3. The difference between pixels is amplified by the weight coefficient in the filter. The Scharr in the x-direction and y-direction is defined as follows:

[0106]

[0107] 3) Use the two Scharr operators in the previous step to perform convolution operations on the brightness component V and obtain two matrices g of the same size as the original image x and g y , the specific calculation formula is as follows, where G x () represents the convolution operation on the image, and the convolution kernel is G x , G y () represents the convolution operation on the image, and the convolution kernel is G y ;

[0108] g x (x,y)=G x (V(x,y)),g y (x,y)=G y (V(x,y)) (12)

[0109] 4) In an image, the areas with light and dark changes and the edges of objects usually have larger gradients, while the areas with less obvious light and dark changes and the non-edges of objects usually have smaller gradients. Considering that areas with less obvious light and dark changes and the non-edges of objects should be more strongly enhanced during image enhancement, the enhancement weight map of gradient information is obtained according to the following formula, where σ represents

[0110] A small constant to avoid the gradient being zero;

[0111]

[0112] 5) The final enhancement weight map is obtained by combining the brightness component V and the re weight map obtained in the previous step and performing a power law transformation, where β = 255, which is used to normalize the brightness component V to between (0, 1), and α is an empirical parameter, which is set to 0.5 in the present invention;

[0113]

[0114] 6) The gradient detection local enhancement module is used to further enhance the image after global tone mapping, which can restore the image detail information. The image after local enhancement is obtained by the following formula:

[0115]

[0116] Specifically, the glare suppression module:

[0117] 1) When performing low-light enhancement, existing algorithms often inappropriately enhance parts of the image with higher brightness (such as car lights and street lights), making the surrounding areas too bright, resulting in a poor visual experience. The present invention uses a linear expansion algorithm to suppress glare;

[0118] 2) Weber's theorem states that the difference in light intensity that people can perceive is related to the intensity of the background light.

[0119] Regions with large change amplitudes are more likely to attract the attention of the human eye. Based on this characteristic, when processing images, the low grayscale and high grayscale regions with narrower amplitudes can be stretched, and the intermediate grayscale regions can be compressed. The present invention adopts a linear expansion algorithm that matches the human eye, as shown in the following formula, where L e Indicates that after the above two modules are enhanced

[0120] The image after L o represents the original image, a1 and a2 represent the corresponding weights;

[0121] L p =a1L o (x,y)+a2L e (x,y) (16)

[0122] L p Represents the result obtained after the linear expansion algorithm;

[0123] 3) The calculation of weights a1 and a2 is as follows. To achieve the desired effect, the weight of darker pixels should be larger in brighter scenes, and the weight of darker pixels should be smaller in darker scenes. The threshold T = 256-V ave , a2=1-a1, where V ave It represents the average value of the image brightness. This method can achieve the purpose of adaptively calculating the thresholds of different images, improve the flexibility of the linear expansion algorithm and make the image achieve better visual effects.

[0124]

[0125] 4) After the image is processed by the glare suppression module, the brighter areas in the image can be suppressed, but there will be a certain degree of contrast reduction and unnatural colors. Since the basic principle of the CLAHE algorithm is to split the image into multiple sub-blocks for contrast-limited histogram equalization, it mainly considers the local information of the image and will not affect the enhancement effect of darker areas due to uneven image lighting. Therefore, the CLAHE algorithm is used to further improve the image contrast while retaining the glare suppression effect.

[0126] 5) In the present invention, the CLAHE algorithm is not directly applied to the three RGB channels of the enhanced image respectively. Instead, the image is first converted from the RGB space to the XYZ space, then from the XYZ space to the Lab space, and then the CLAHE algorithm is applied to the L component. The formula used for the space conversion is as follows (18):

[0127] (19), (20) as shown:

[0128]

[0129]

[0130] 6) Apply the CLAHE algorithm to the L component obtained in the previous step. In the present invention, the contrast l in the CLAHE algorithm is set to 1, and the size of the divided sub-image is set to 10×10. After the CLAHE algorithm is enhanced, the final output image is obtained, as shown below, where L Le Represents the L component after Le is converted to Lab space;

[0131] L out =CLAHE(L Le ,l,size) (21)

[0132] 7) The L component obtained after contrast processing of the image and other unprocessed components are combined and converted back to the RGB space to obtain the final enhanced image of the present invention.

[0133] ①Advantages and uses of this product.

[0134] like Figure 4 As shown, compared with the existing low-light image enhancement methods, the present invention combines low-light image enhancement with glare suppression, taking into account various night scenes and lighting conditions, and can not only effectively restore the details of the darker areas in the image, but also will not amplify the bright spots and halos that already exist in the picture. It can suppress the brighter areas in the image to a certain extent and achieve brightness balance for the entire image. In addition, the present invention has the characteristics of high real-time performance while having a good enhancement effect, and balances the enhancement effect and enhancement efficiency, and can guarantee good real-time performance at a variety of image resolution sizes. In summary, the present invention can take into account the restoration tasks of multiple lighting scenes at the same time, has a good enhancement effect, and can guarantee good real-time performance and be applied in practical applications.

[0135] ②The process or method of using the product.

[0136] As an image processing algorithm, the present invention can be combined with existing equipment and deployed on its computing backend to perform secondary processing and enhancement on images. Specifically, it can be used for:

[0137] Reconnaissance UAV: ​​It can be used for reconnaissance UAV, which can ensure the visibility of images in dark environments and improve the reconnaissance efficiency and effect.

[0138] Smart car: It can be used for smart self-driving cars, which can perform self-driving tasks at night or in low-light conditions to ensure the accuracy and safety of self-driving.

[0139] Individual equipment: It can be used for individual combat equipment to improve the soldier's night combat capability, reduce dependence on light sources and ensure concealment.

[0140] Special vehicles: can be used for special vehicles to ensure that special vehicles can perform corresponding tasks at night and in dim light conditions, thereby improving efficiency and response speed.

[0141] Security monitoring: It can be used for security monitoring cameras to conduct video monitoring in a variety of occasions, improve image visual details, and enhance image visual effects.

Claims

1. A glare suppression low-light image enhancement method based on global mapping and local tuning, characterized in that: The following steps are involved: Step S1: Perform preliminary enhancement on the original image, perform global tone mapping through an improved algorithm parameter optimization strategy to restore certain details and information of the original image, and adjust the color saturation and contrast of the image; Step S2: performing image region gradient detection on the original image to calculate the region with more drastic changes in brightness and darkness, and further enhancing the image adaptively through a corresponding algorithm; Step S3: The images before and after enhancement are superimposed by a linear extension algorithm and a pixel-level weight adaptive algorithm, thereby suppressing the glare in the image, and the algorithm is used to adaptively adjust the contrast and color of the image after glare suppression, so that the image conforms to the visual habits of the human eye.

2. The method for low-light image enhancement based on global mapping and local tuning for glare suppression according to claim 1, characterized in that: The step S1 specifically includes the following steps: Step S11: The specific calculation formula for converting the original image from the RGB space to the HSV space is as follows: V=max(R,G,B) (1) Among them, H represents the hue component of the HSV space, which refers to the type or hue of the color, S represents the saturation component of the HSV space, which refers to the vividness of the color, and V represents the brightness component of the HSV space, which refers to the brightness of the color; Step S12: Calculate the logarithmic mean value Y of the brightness using the V component calculated in step S11 ave : Where MN represents the product of the height and width of the original image, σ represents a constant, x represents the horizontal coordinate of the pixel, and y represents the vertical coordinate of the pixel; Step S13: Calculate the standard deviation δ of all pixels in the V component using formula (5), where v(i, j) represents a pixel on the V component, and μ represents the pixel mean of the original image; Step S14: according to different pixel standard deviations, the input original image is classified according to formula (6), and the input original image is classified into three categories: an image with non-uniform illumination, an image with uniform illumination, and an image with brightness difference between the two; Where δ represents the standard deviation of the input original image pixels, δ1 = 0.17, δ2 = 0.09; Step S15: Different L ave Different adjustment functions are used to perform adjustments to enhance the curvature of the global tone mapping function curve, thereby enhancing the adaptability to various dark scenes with different brightness conditions.

3. The method for low-light image enhancement based on global mapping and local tuning for glare suppression according to claim 2, characterized in that: The step S1 further comprises: Step S16: The optimization adjustment strategy of the brightness logarithm mean can be specifically represented by the following piecewise function, where Y c Represents the logarithmic mean of brightness after adjustment by the adjustment function; Step S17: Introduce parameter P1 to adjust Y c and V max , the adjustment formula is as follows: Y cln =ln(P1Y c +1) V max ln =ln(P1V max +1) (8) where Y cln It represents the logarithmic mean of brightness obtained after the above adjustment, V maxln represents the maximum brightness obtained after the above adjustment, and P1 represents an empirical parameter used to adjust the representation capability of the enhancement curve to improve the final effect of the present invention; Step S18: Using the obtained Y cln and V maxln Substitute the following formula for calculation and introduce the adjustment parameter P2, where L e Represents the result obtained after preliminary enhancement of global tone mapping; Step S19: Calculate the enhancement curve cutoff threshold V close to the value obtained by the P1 and P2 parameters using the Newton iteration algorithm. cut , where the loop parameter of the Newton iteration method is set to 20 in the present invention, and only the pixel part smaller than the threshold value will be enhanced by the enhancement curve, so we get:

4. The method for low-light image enhancement based on global mapping and local tuning for glare suppression according to claim 1, characterized in that: The step S2 further comprises: Step S21: using the Scharr operator to perform a convolution operation on the brightness component V obtained in the global tone mapping module, thereby extracting the gradient information at each pixel and obtaining a gradient weight information map in the x direction and the y direction; Step S22: The size of the Scharr operator is 3×3. The difference between pixels is magnified by the weight coefficient in the filter. The Scharr in the x-direction and the y-direction is defined as follows: Step S23: Use the two Scharr operators in step S22 to perform convolution operations on the brightness component V, and obtain two matrices g with the same size as the original image. x and g y The specific calculation formula is as follows, where G x () represents the convolution operation on the image, and the convolution kernel is G x , G y () represents the convolution operation on the image, and the convolution kernel is G y ; g x (x,y)=G x (V(x,y)),g y (x,y)=G y (V(x,y)) (12); Step S24: Obtain an enhanced weight map of the gradient information according to the following formula, where σ represents a very small constant to avoid the situation where the gradient is 0; Step S25: The final enhancement weight map is obtained by combining the brightness component V and the re weight map obtained in the previous step and performing a power law transformation, where β = 255, which is used to normalize the brightness component V to between (0, 1), and α is an empirical parameter; Step S26: The gradient detection local enhancement module is used to further enhance the image after global tone mapping, and can restore image detail information. The image after local enhancement is obtained by the following formula:

5. The method for low-light image enhancement based on global mapping and local optimization for glare suppression according to claim 1, characterized in that: The step S3 specifically includes: Step S31: using a linear expansion algorithm to suppress glare; Step S32: Using a linear expansion algorithm that matches the human eye, as shown in the following formula, where L e It represents the image enhanced by the above two modules, L o represents the original image, a1 and a2 represent the corresponding weights; <h2 style=";text-align:left;direction:ltr">L<h2 style=";text-align:left;direction:ltr"> p <h2 style=";text-align:left;direction:ltr"> =a1L<h2 style=";text-align:left;direction:ltr"> o <h2 style=";text-align:left;direction:ltr"> (x,y)+a2L<h2 style=";text-align:left;direction:ltr"> e <h2 style=";text-align:left;direction:ltr"> (x,y) (16) L p Represents the result obtained after the linear expansion algorithm; Step S33: The weights a1 and a2 in the linear extension algorithm have different values ​​at different brightness positions. In brighter pixel areas, the weight of the original image is higher, and in darker pixel areas, the weight of the original image is lower; wherein the threshold T = 256-V ave , a2=1-a1, where V ave Indicates the average value of image brightness; Step S34: using the CLAHE algorithm to further improve the image contrast while retaining the glare suppression effect; Step S35: Convert the image from RGB space to XYZ space, then from XYZ space to Lab space, and then apply the CLAHE algorithm to the L component. The formulas used for space conversion are shown in (18), (19), and (20):

6. The method for low-light image enhancement based on global mapping and local tuning for glare suppression according to claim 4, characterized in that: The step S3 specifically includes: Step S36: Apply the CLAHE algorithm to the L component obtained in step S35, set the contrast l in the CLAHE algorithm to 1, and set the size of the divided sub-images to 10×10. After the CLAHE algorithm is enhanced, the final output image is obtained, as shown below, where L Le Represents the L component after Le is converted to Lab space; L out =CLAHE(L Le ,l,size) (21) Step S37: The L component obtained after contrast processing of the image and other unprocessed components are combined and converted back to the RGB space to obtain the final enhanced image of the present invention.

7. A glare suppression low-light image enhancement system based on global mapping and local tuning, characterized in that: A glare suppression low-illuminance image enhancement method based on global mapping and local tuning as described in any one of claims 1 to 6 is executed; and also includes a global tone mapping enhancement module, a gradient detection local enhancement module and a glare suppression module.

8. The glare suppression low-light image enhancement system based on global mapping and local tuning according to claim 7, characterized in that: The global tone mapping enhancement module is responsible for performing preliminary enhancement on the original image, performing global tone mapping through an improved algorithm parameter optimization strategy to restore certain details and information of the original image, and adjusting the color saturation and contrast of the image.

9. The glare suppression low-light image enhancement system based on global mapping and local tuning according to claim 7, characterized in that: The gradient detection local enhancement module performs image region gradient detection on the original image to calculate the region with more drastic changes in brightness and darkness, and further enhances the image adaptively through a corresponding algorithm.

10. The glare suppression low illumination image enhancement system based on global mapping and local tuning according to claim 7, characterized in that: The glare suppression module uses a linear expansion algorithm and a pixel-level weight adaptive algorithm to superimpose the images before and after enhancement, thereby suppressing the glare in the image, and uses an algorithm to adaptively adjust the contrast and color in the image after glare suppression to make the image conform to the visual habits of the human eye.

Images