A low-illumination image enhancement method based on global gray scale lifting and local gray scale compensation

Abstract

The application discloses a low-illumination image enhancement method based on global gray scale promotion and local gray scale compensation, and belongs to the technical field of image processing.The application solves the problem that the existing image enhancement method is prone to compressing or losing the details and texture information of an image.The main technical scheme of the method is as follows: step one, performing brightness normalization processing on a low-illumination image to obtain an image after brightness normalization processing;step two, performing contrast stretching on the image after brightness normalization processing to obtain an image after contrast stretching, and obtaining an image after global gray scale promotion after processing the image after contrast stretching; and step three, performing local gray scale compensation on the image after global gray scale promotion to obtain an enhanced image after local gray scale compensation.The method can be applied to low-illumination image enhancement.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology, specifically relating to a low-light image enhancement method based on global grayscale enhancement and local grayscale compensation. Background Technology

[0002] Low-light image enhancement technology refers to techniques that improve the brightness, contrast, and sharpness of images under insufficient lighting conditions. Currently, low-light image enhancement technology has been widely applied and researched in fields such as computer vision, image processing, security monitoring, and night vision devices.

[0003] Traditional low-light image enhancement methods mainly include histogram equalization-based methods, spatial filtering-based methods, and Retinex-based methods. However, while improving the visibility of low-light images, these methods tend to compress or lose image details and texture information, resulting in a failure to guarantee the detail and sharpness of the enhanced image. Summary of the Invention

[0004] The purpose of this invention is to address the problem that existing image enhancement methods tend to compress or lose image details and texture information, and to propose a low-light image enhancement method based on global grayscale enhancement and local grayscale compensation.

[0005] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows:

[0006] A low-light image enhancement method based on global grayscale enhancement and local grayscale compensation, the method specifically includes the following steps:

[0007] Step 1: Perform brightness normalization processing on the low-light image to obtain the brightness-normalized image I;

[0008] Step 2: Perform contrast stretching on image I after brightness normalization to obtain the contrast-stretched image. Contrast-stretched image After processing, the image L with global grayscale enhancement is obtained;

[0009] Step 3: Perform local grayscale compensation on the globally enhanced image L to obtain the enhanced image after local grayscale compensation.

[0010] Furthermore, the specific process of step two is as follows:

[0011]

[0012] Where I(x,y) is the brightness of pixel (x,y) in the image after brightness normalization, and ln is the logarithm to the base e. 0.2987 (x,y) represents I(x,y) raised to the power of 0.2987. It is the brightness of the pixel (x,y) in the image after contrast stretching;

[0013]

[0014] Where L(x,y) is the brightness of pixel (x,y) in the image after global grayscale enhancement. It is the minimum brightness of a pixel in the contrast-stretched image. It is the maximum brightness of a pixel in the image after contrast stretching.

[0015] Furthermore, the specific process of step three is as follows:

[0016]

[0017] in, R(x,y) is the brightness of pixel (x,y) in the enhanced image after local gray-level compensation, and R(x,y) is the local gray-level compensation coefficient of pixel (x,y).

[0018] Furthermore, the calculation method for the local grayscale compensation coefficient R(x,y) is as follows:

[0019]

[0020] Wherein, min(L(x+i,y+j),m(x,y)) represents taking the smaller value between L(x+i,y+j) and m(x,y), and max(L(x+i,y+j),m(x,y)) represents taking the larger value between L(x+i,y+j) and m(x,y). L(x+i,y+j) is the brightness of pixel (x+i,y+j) in the image after global grayscale enhancement, and m(x,y) is an intermediate variable.

[0021] Furthermore, the expression for the intermediate variable m(x,y) is shown in equation (5):

[0022]

[0023] The beneficial effects of this invention are:

[0024] The method of this invention first designs a global grayscale enhancement function to increase the brightness of low-light pixels, thereby improving the visibility of low-light areas in the image; secondly, it designs a local grayscale compensation method to compensate for the image details and texture information that are compressed or lost during the global grayscale enhancement process, making the enhanced image clearer and brighter with richer details, thus ensuring the detail and clarity of the enhanced image.

[0025] Experimental results show that the low-light image enhancement method designed in this invention can make the enhanced image have a better subjective visual effect. Attached Figure Description

[0026] Figure 1 This is a flowchart of the method of the present invention;

[0027] Figure 2 It is a global grayscale enhancement curve;

[0028] Figure 3 It is an underwater image;

[0029] Figure 4 yes Figure 3 Enhanced images;

[0030] Figure 5 It is an indoor image;

[0031] Figure 6 yes Figure 5 Enhanced images;

[0032] Figure 7 It is a library image;

[0033] Figure 8 yes Figure 7 Enhanced image. Detailed Implementation

[0034] Specific Implementation Method 1: Combination Figure 1 This embodiment describes a low-light image enhancement method based on global grayscale enhancement and local grayscale compensation. The method specifically includes the following steps:

[0035] Step 1: Perform brightness normalization processing on the low-light image to obtain the brightness-normalized image I;

[0036] Step 2: Perform contrast stretching on image I after brightness normalization to obtain the contrast-stretched image. Contrast-stretched image After processing, the image L with global grayscale enhancement is obtained;

[0037] Step 3: Perform local grayscale compensation on the globally enhanced image L to obtain the enhanced image after local grayscale compensation.

[0038] The method described in this embodiment can enhance low-light images to produce enhanced images with better brightness, contrast, and detail.

[0039] Specific Implementation Method Two: Combining Figure 2 This embodiment is described below. The difference between this embodiment and specific embodiment one is that the specific process of step two is as follows:

[0040]

[0041] Where I(x,y) is the brightness of pixel (x,y) in the image after brightness normalization, and ln is the logarithm to the base e. 0.2987 (x,y) represents I(x,y) raised to the power of 0.2987. It is the brightness of the pixel (x,y) in the image after contrast stretching;

[0042]

[0043] Where L(x,y) is the brightness of pixel (x,y) in the image after global grayscale enhancement. It is the minimum brightness of a pixel in the contrast-stretched image. It is the maximum brightness of a pixel in the image after contrast stretching.

[0044] The other steps and parameters are the same as in Specific Implementation Method 1.

[0045] Specific Implementation Method Three: This implementation method differs from Specific Implementation Method One or Two in that the specific process of step three is as follows:

[0046]

[0047] in, R(x,y) is the brightness of pixel (x,y) in the enhanced image after local gray-level compensation, and R(x,y) is the local gray-level compensation coefficient of pixel (x,y).

[0048] Other steps and parameters are the same as in specific implementation method one or two.

[0049] Specific Implementation Method Four: This implementation method differs from Specific Implementation Methods One to Three in that the calculation method for the local grayscale compensation coefficient R(x,y) is as follows:

[0050]

[0051] In this context, min(L(x+i,y+j),m(x,y)) represents taking the smaller value between L(x+i,y+j) and m(x,y), and max(L(x+i,y+j),m(x,y)) represents taking the larger value between L(x+i,y+j) and m(x,y). L(x+i,y+j) is the brightness of pixel (x+i,y+j) in the globally grayscale-enhanced image, and m(x,y) is an intermediate variable. The value range of i is -2, -1, 0, 1, 2, and the value range of j is -2, -1, 0, 1, 2.

[0052] The other steps and parameters are the same as those in one of the specific implementation methods one to three.

[0053] Specific Implementation Method Five: This implementation method differs from Specific Implementation Methods One to Four in that the expression for the intermediate variable m(x,y) is shown in Equation (5):

[0054]

[0055] The other steps and parameters are the same as those in one of the specific implementation methods one to four.

[0056] Experimental Results and Analysis

[0057] The experiment used a desktop computer to simulate the low-light image enhancement method designed in this invention. The configuration was as follows: Intel Core i7-10700K CPU with a clock speed of 3.80GHz; 32GB of memory with a frequency of 3200MHz; Nvidia GeForce RTX 3070 graphics card; Windows 10 operating system; and Matlab R2022a simulation software. The experimental results are as follows. Figures 3 to 8 As shown.

[0058] from Figures 3 to 8 It is evident that the enhanced image offers a better subjective visual effect compared to the original low-light image. The original low-light image often suffers from blurred details, excessive darkness in many areas, and low image contrast, making it difficult to observe and recognize clearly with the naked eye. The method designed in this invention can make the image clearer and brighter, with richer details, thereby improving image quality and readability. Therefore, processing low-light images using the low-light image enhancement method designed in this invention can achieve a better subjective visual effect.

[0059] The above examples of the present invention are merely illustrative of the computational model and process of the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is impossible to exhaustively list all possible implementations here. Any obvious variations or modifications derived from the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A low-light image enhancement method based on global gray scale lifting and local gray scale compensation, characterized in that, The method specifically comprises the following steps: Step one, brightness normalization processing is conducted on the low-illumination image to obtain a brightness normalized image ; Step two, normalize the image by brightness Step three, stretch the contrast of the image Step three, stretch the contrast of the image Step four, get the image with global gray scale enhancement Step four, get the image with global gray scale enhancement Step three, the image after global gray scale enhancement carries out local gray scale compensation, and obtains an enhanced image after local gray scale compensation; The specific process of the step three is: (3) wherein, is the luminance of the pixel point in the enhanced image after local gray compensation, is the local gray compensation coefficient of the pixel point , and is the luminance of the pixel point in the image after global gray enhancement. The local gray scale compensation coefficient The calculation method is: (4) wherein, represents taking the smaller value from and , represents taking the larger value from and , is the luminance of the pixel point in the globally gray scale enhanced image, is the luminance of the pixel point in the luminance normalized image, is an intermediate variable; The intermediate variable The expression of the intermediate variable is given by equation (5): (5)。 2. The low-light image enhancement method based on global gray scale lifting and local gray scale compensation according to claim 1, characterized in that, The specific process of the step two is: (1) wherein, is the base-e logarithm, represents 0.2987th power of, is the luminance of the pixel point in the contrast-stretched image; (2) wherein, is the minimum luminance of a pixel in the contrast-stretched image, is the maximum luminance of a pixel in the contrast-stretched image.

Images