Coal mine underground image enhancement method, system, computer device and storage medium

Abstract

The present application belongs to the technical field of image enhancement in coal mine, and specifically discloses a coal mine image enhancement method, system, computer device and storage medium. Firstly, a glow imaging degradation model is established to obtain a glow image, and a layer separation method is used to remove the glow to obtain a de-glow image; then, on the basis of a low-illumination enhancement SRLLIE method, a target function of the low-illumination enhancement SRLLIE method is improved according to the de-glow image, the improved target function is iteratively solved, an illumination graph with suppressed overexposure and retained structural details and a de-noised reflection graph are obtained; then, the brightness of the illumination graph is adjusted by using an S-shaped gamma correction function to obtain an optimized illumination graph; finally, the optimized illumination graph and the reflection graph are point multiplied according to the Retinex theory to obtain an enhanced image. The present application can not only effectively improve the brightness of the image and suppress over-enhancement, but also restore the image detail information and improve the visual effect of the image.

Description

Technical Field

[0001] This invention belongs to the field of underground coal mine image enhancement technology, specifically relating to an underground coal mine image enhancement method, system, computer equipment, and storage medium. Background Technology

[0002] Underground video images serve as a crucial carrier of information in underground mines. Extensive use of image and video technologies throughout the entire coal mining process provides a solid data foundation for intelligent monitoring of underground coal mines. However, due to the absorption and multiple scattering of light emitted by artificial light sources by coal dust particles floating underground, glow phenomena occur in the images, leading to uneven illumination. Furthermore, insufficient light and harsh imaging environments in underground coal mines result in low illumination and high noise levels, degrading the video images and hindering the identification of critical information. This impedes real-time monitoring of mine workers, equipment operation, and accurate decision-making by intelligent systems, thus limiting the application of digital video technology in coal mine production.

[0003] Ying Dongjie et al. used wavelet transform technology to decompose and reconstruct underground coal mine images, enhancing the image's detail information. Fan Lingyun et al., on the other hand, performed histogram equalization and wavelet transform on underground images, followed by matching processing and then inverse wavelet transform to enhance the image. However, these wavelet transform-based image enhancement methods did not adequately address the enhancement of low-frequency information, resulting in ineffective image information display and failing to significantly improve overall image brightness.

[0004] Zhi et al. used the concept of a dual gamma function, fine-tuning it based on the distribution characteristics of the illumination map, to improve the gray level of low-light areas and reduce the gray level influence of local bright areas. Yu et al. applied arctangent hyperbolas and optimized hyperbolic tangent profiles to perform tone component mapping and enhancement processing on low-light images. Furthermore, Park et al. proposed a two-layer autoencoder network model based on Retinex theory (an image processing theory), which combines stacked autoencoders and convolutional autoencoders to achieve low-light enhancement and noise reduction. Kimmel et al., utilizing prior assumptions, proposed a Retinex image enhancement method based on a variational framework, transforming the illumination estimation problem into an optimal quadratic optimization problem; although highly complex, it yields significant results. Elad et al. proposed a non-iterative Retinex image enhancement method that effectively controls edges in images and suppresses noise in dark areas.

[0005] Although the above methods have achieved certain results in image enhancement, the processing effect of existing image enhancement methods is not ideal due to problems such as low and uneven illumination and a lot of noise in underground coal mine images. Summary of the Invention

[0006] To address the problems of low and uneven illumination and high noise in underground coal mine images, this invention proposes an image enhancement method for underground coal mines. This method improves the objective function of the low-illuminance enhancement SRLLIE method by using deglow image processing. It can effectively enhance and denoise low-illuminance images while suppressing overexposure, resulting in enhanced images that effectively improve brightness, suppress over-enhancement, and recover image detail information, thus improving the visual effect of the image.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] Step 1. Establish a glow imaging degradation model to obtain a glow image, use a layer separation method to remove the glow, suppress overexposure, and obtain a degloss image;

[0009] Step 2. Based on the low-light enhancement SRLLIE method, the objective function of the low-light enhancement SRLLIE method is improved according to the deglow image. The improved objective function is solved iteratively to obtain the illumination map that suppresses overexposure and preserves structural details and the denoised reflection map.

[0010] Step 3. Adjust the brightness of the illuminance map using the S-type gamma correction function to obtain the optimized illuminance map;

[0011] Step 4. According to Retinex theory, multiply the optimized illuminance map by the reflectance map to obtain the enhanced image.

[0012] Furthermore, based on existing methods for enhancing images in underground coal mines, this invention also proposes a corresponding image enhancement system for underground coal mines, the technical solution of which is as follows:

[0013] A coal mine underground image enhancement system, comprising:

[0014] The image deglow module is used to establish a glow imaging degradation model to obtain a glow image. It uses a layer separation method to remove the glow, avoids overexposure problems, and obtains a deglow image.

[0015] The improved SRLLIE module is used to improve the objective function of the low-light enhancement SRLLIE method based on the deglow image, and iteratively solve the improved objective function to obtain an illumination map that suppresses overexposure and preserves structural details, and a denoised reflectance map.

[0016] The illuminance map optimization module is used to adjust the brightness of the illuminance map using the S-shaped gamma correction function to obtain an optimized illuminance map.

[0017] The image enhancement module is used to multiply the optimized illuminance map and reflectance map by a dot product according to Retinex theory to obtain the enhanced image.

[0018] Furthermore, based on the aforementioned method for enhancing images in underground coal mines, this invention also proposes a computer device comprising a memory and one or more processors;

[0019] The memory stores executable code, and when the processor executes the executable code, it implements the steps of the coal mine underground image enhancement method described above.

[0020] Furthermore, based on the aforementioned method for enhancing underground coal mine images, this invention also proposes a computer-readable storage medium storing a program thereon; when executed by a processor, this program is used to implement the steps of the aforementioned method for enhancing underground coal mine images.

[0021] The present invention has the following advantages:

[0022] As described above, this invention discloses an image enhancement method for underground coal mines. The method first establishes a glow imaging degradation model to obtain a glow image. A layer separation method is used to subtract the glow image from the input image to remove the glow, resulting in a degloss image that suppresses overexposure and improves image clarity and recognizability. Next, based on the existing low-light enhancement SRLLIE method, the objective function is improved according to the degloss image. The improved objective function is then iteratively solved using the alternating direction multiplier method to obtain an illuminance map that suppresses overexposure and preserves structural details, and a denoised reflectance map. This avoids overexposure, improves the uneven illumination problem in underground coal mine images, and reduces noise interference. Then, the brightness of the illuminance map is adjusted using the sigmoid gamma correction function. The resulting optimized illuminance map improves the brightness of dark areas while avoiding over-enhancing overly bright areas. Finally, according to Retinex theory, the optimized illuminance map and reflectance map are multiplied to obtain the enhanced image. The method of this invention can not only effectively improve the brightness of the image and suppress excessive enhancement, but also restore the image detail information, improve the visual effect of the image, and effectively solve the problems of low and uneven illumination and excessive noise in underground coal mine images. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the underground coal mine image enhancement method in an embodiment of the present invention.

[0024] Figure 2 This is a flowchart of the underground coal mine image enhancement method in an embodiment of the present invention. Detailed Implementation

[0025] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0026] Example 1

[0027] To address the problems of low and uneven illumination and high noise in underground coal mine images, this embodiment describes an image enhancement method for underground coal mines. The general inventive concept of this method is as follows: First, glow is removed using a light source (glow) model to avoid overexposure and obtain a deglowed image. Then, based on the low illumination enhancement SRLLIE algorithm, the objective function is improved according to the deglowed image, and iteratively solved using the alternating direction multiplier method to obtain an illuminance map that suppresses overexposure and preserves structural details, and a denoised reflectance map. Next, the brightness of the illuminance map is adjusted using the sigmoid gamma correction function to obtain an optimized illuminance map. Finally, according to Retinex theory, the optimized illuminance map and the reflectance map are multiplied to obtain the enhanced image.

[0028] like Figures 1 to 2 As shown, the coal mine underground image enhancement method in this embodiment includes the following steps:

[0029] Step 1. Establish a glow imaging degradation model to obtain a glow image, use a layer separation method to remove the glow, suppress overexposure, and obtain a degloss image.

[0030] Light from a light source undergoes multiple scattering by fine particles, forming a glow. This glow can cause overexposure near the light source in the image. This glow is modeled as an atmospheric point spread function. Based on this, a glow imaging degradation model is established as follows:

[0031] I = J + L a *APSF (1)

[0032] Where I is the input image, J represents the degloss image; L a *APSF is a glow image G, L a The light source is represented by *, which indicates a convolution operation, and APSF represents the atmospheric point spread function.

[0033] Glow images are obtained based on the glow imaging degradation model of formula (1), which are used to suppress overexposure and improve image clarity and recognizability.

[0034] Because light is scattered multiple times by tiny particles, the brightness of the glow gradually and smoothly decreases, and the gradient histogram of the glow exhibits a "short-tailed" distribution. This property is utilized to remove the glow using a layer separation method. The objective function of the layer separation is:

[0035]

[0036] Where E(J) represents the objective function for layer separation, f 1,2f1 is a first-order derivative filter in the horizontal and vertical directions, f2 is a second-order Laplace filter; ρ(s) is a robust function that can remove glow and retain information with large gradients in the input image, ρ(s) = min(s) 2 ,τ) where s is J(x,y)*f 1,2 τ is a fixed parameter; λ is the smoothness weight parameter, J r J represents the r channel of the deglow image. g J represents the g channel of the deglow image. b This represents the b channel of the degloss image.

[0037] The inequality constraints in formula (2) ensure that the solution J of the objective function is within an appropriate range (i.e., ensure that J is greater than 0 and less than 1), while the equality constraints in formula (2) can effectively solve the color shift problem.

[0038] Solve the objective function of the above formula (2) to obtain the degloss image.

[0039] Because the air in underground coal mines contains a large amount of coal dust particles, and underground lighting relies solely on artificial light sources, these suspended particles cause multiple scattering of the light emitted by the light source, resulting in glow phenomena near the image light source and leading to overexposure problems in the image. Therefore, this invention utilizes a glow model to remove glow, suppress overexposure, and improve image clarity and recognizability.

[0040] Step 2. Based on the low-light enhancement SRLLIE method, the objective function of the low-light enhancement SRLLIE method is improved according to the deglow image obtained in Step 1. The improved objective function is solved iteratively to obtain an illumination map that suppresses overexposure and preserves structural details, and a denoised reflection map, thereby improving the problem of uneven illumination in coal mine images, avoiding overexposure, and reducing noise interference.

[0041] SRLLIE is a high-performance low-light image enhancement method; however, when processing images with uneven illumination in underground coal mines, it can lead to overexposure near the near-light points. To address this, this invention proposes an improved SRLLIE algorithm for underground coal mine images based on a glow model. Specifically, the input image in the objective function of the original SRLLIE algorithm is replaced with the deglossed image obtained in step 1. The improved objective function formula is as follows:

[0042]

[0043] Where I represents the input image, R represents the reflectance map, L represents the illuminance map, G represents the glow image, and N represents the noise map; β, ω, and δ are coefficients; ||·|| FAnd ||·||1 represent the F-norm and L1 norm, respectively. It is a first-order differential operator, and P is the adjusted gradient of the input image.

[0044] Based on the improved objective function of the low-light enhancement SRLLIE method, the solution is obtained iteratively using the alternating direction multiplier method. The specific solution process is as follows:

[0045] First, we introduce an auxiliary variable T to replace the second term of the objective function in formula (3). The objective function is transformed into the following form:

[0046]

[0047] By introducing the Lagrange multiplier Z to eliminate the equality constraints, we obtain the augmented Lagrange function:

[0048]

[0049] in, A is Z, B is <*,*> denotes the matrix inner product, and μ denotes the penalty scalar.

[0050] The first term on the right side of equation (5) The term is the fidelity term; the second term on the right side of equation (5), ||T||1, is the smoothing term, taking into account the smoothness of the illuminance map L; the third term on the right side of equation (5) is... Constrain the fidelity between the gradient of the reflection map and the gradient of the input image, and enhance the structural information of the reflection map; the fourth term on the right side of equation (5) Limit the overall intensity of noise. The fidelity term in the noise map is used to ensure the accuracy of the model, and it is expected that the estimated illuminance map, reflectance map, and noise map can accurately reconstruct the input image.

[0051] The objective function is solved iteratively step by step. The following is the calculation process of the k-th iteration solution for each subproblem:

[0052] R subproblem: Ignoring the terms in formula (5) that are irrelevant to R, we obtain the following objective function:

[0053]

[0054] Among them, L (k) Let N represent the result of the k-th iteration of L. (k) Let N represent the result of the k-th iteration; by restating the objective function in formula (6), it becomes a classic least squares problem:

[0055]

[0056] Where l, r, n, i, g, and p are the vectorizations of L, R, N, I, G, and P, respectively. Let l represent a diagonal matrix with elements l. express The result of the kth iteration, n (k) This represents the result of the k-th iteration of n.

[0057] By differentiating r in formula (7), treating all variables in formula (7) except r as constants, and setting the derivative with respect to r to 0, we obtain the following formula:

[0058]

[0059] Where D is the discrete gradient operator, f(x) = x T x, r (k+1) This represents the result of the (k+1)th iteration of r.

[0060] L-subproblem: Ignoring the terms in formula (5) that are irrelevant to L, we obtain the following objective function:

[0061]

[0062] Among them, R (k+1) Z represents the result of the (k+1)th iteration of R. (k) Let T represent the result of the k-th iteration of Z. (k) Let T represent the result of the k-th iteration; the solution of the (k+1)-th iteration of L is as follows:

[0063]

[0064] Where t is the vectorization of T, and z is the vectorization of Z. express The result of the (k+1)th iteration, Let t represent a diagonal matrix with elements l. (k) Let z represent the result of the k-th iteration of t. (k) This represents the result of the k-th iteration of z.

[0065] N subproblem: Ignoring the terms in formula (5) that are irrelevant to N, we obtain the following objective function:

[0066]

[0067] The solution for the (k+1)th iteration of N is:

[0068]

[0069] Subproblem T: Ignoring the terms in formula (5) that are irrelevant to T, we obtain the following objective function:

[0070]

[0071] Among them, Z (k) Let Z represent the result of the k-th iteration. The solution of T for the (k+1)-th iteration is:

[0072]

[0073] Among them, S ε (x)=sign(x)max(|x|-ε,0), ε is The calculation is performed element-wise; μ (k) This represents the result of the k-th iteration of μ.

[0074] The auxiliary matrix Z and the penalty scalar μ are updated in the following way:

[0075]

[0076] Where h is a constant and h > 1. Only when R (k) and R (k+1) The difference between or L (k) and L (k+1) The entire iteration will stop when the difference between the values ​​is less than a preset threshold or when the maximum number of iterations is reached.

[0077] In this embodiment, h is, for example, taken as 1.5, and the threshold value is preset to 10. -3 The maximum number of iterations is set to 30.

[0078] After the iteration, an illumination map that suppresses overexposure and preserves structural details and a denoised reflectance map are obtained.

[0079] This invention improves the objective function of the original SRLLIE algorithm based on the deglow image, thereby obtaining an illumination map that suppresses overexposure and preserves structural details, and a denoised reflection map. This avoids overexposure, improves the problem of uneven illumination in underground coal mine images, and also reduces noise interference.

[0080] Step 3. Adjust the brightness of the illuminance map using the S-type gamma correction function to obtain an optimized illuminance map.

[0081] The illuminance map obtained through iterative solution is still low in brightness, therefore the brightness of the illuminance map is adjusted using the sigmoid gamma correction function:

[0082]

[0083] Where L′ is the optimized illuminance map and γ is the brightness adjustment factor. The optimized illuminance map obtained using the sigmoid gamma correction function can improve the brightness of dark areas while avoiding over-enhancing of overly bright areas.

[0084] Step 4. According to Retinex theory, multiply the optimized illuminance map by the reflectance map to obtain the enhanced image.

[0085] According to Retinex theory, the enhanced image is obtained by multiplying the optimized illuminance map by the reflectance map:

[0086] I′=R·L′ (17)

[0087] Where I′ represents the enhanced image.

[0088] In addition, to verify the effectiveness of the image enhancement method proposed in this invention, the following specific experiments are also given:

[0089] Because the illuminance map obtained by the SRLLIE image enhancement method contains glow near the light source, this will ultimately lead to overexposure in the enhanced image, reducing the visibility of that area. The image enhancement method proposed in this invention improves the original objective function of SRLLIE based on the degloss image, resulting in an illuminance map that reduces glow near the light source, thereby suppressing overexposure.

[0090] The experiment selected three underground coal mine images (I, II, and III) from the dataset and compared the image enhancement method proposed in this invention with the LR3M, BioEnh, and SRLLIE image enhancement methods.

[0091] Among them, the LR3M image enhancement method is an image enhancement method for low-light images that estimates the illumination map and reflectance map based on a low-rank regularized Retinex model. The BioEnh image enhancement method proposes a dual-channel model, namely a structure channel and a detail channel, which is decomposed based on the total variation. The structure channel and detail channel are respectively enhanced for brightness and noise suppression, and finally the two channels are integrated to obtain the enhanced image. The SRLLIE image enhancement method is an image enhancement method that estimates the illumination map and reflectance map based on a robust Retinex model, and then obtains the enhanced image.

[0092] The experimental results show that the image enhancement method proposed in this invention, along with the three comparative image enhancement methods, can enhance coal mine underground images to varying degrees. While the LR3M image enhancement method can improve the brightness of dark areas, it does not eliminate or suppress the glow around the light source, leading to overexposure in areas near the light source. The BioEnh image enhancement method enhances the overall brightness of the image and recovers details in low-light areas, but it also suffers from overexposure near the light source, resulting in severe loss of image texture details. The overexposure problem of the BioEnh image enhancement method is more pronounced compared to other image enhancement methods. The SRLLIE image enhancement method significantly enhances the brightness of low-light images and eliminates noise, but it also leads to overexposure at near-light points, severely reducing the visibility of scenes near the light source.

[0093] Compared with the other three traditional image enhancement methods, the image enhancement method of this invention can effectively improve the brightness of dark areas, restore image details and edge information, and effectively reduce the influence of glow, avoid overexposure in high-brightness areas, thereby enhancing image contrast and improving the overall visual effect.

[0094] Furthermore, to further verify the effectiveness of the image enhancement method proposed in this invention, three no-reference metrics—Information Entropy, Natural Image Quality Evaluation (NIQE), and Average Gradient (AG)—were used to conduct experimental evaluations of the image enhancement method proposed in this invention, as well as the LR3M, BioEnh, and SRLLIE image enhancement methods.

[0095] Information entropy is used to evaluate the information content of an image; a higher information entropy value means the image contains richer visual information. Natural image quality assessment evaluates image quality by analyzing its naturalness and distortion; a lower natural image quality assessment value indicates better visual quality. Average gradient is a key indicator for measuring local detail differences and texture variations in an image, effectively reflecting overall image quality and, to some extent, image sharpness. A higher average gradient value indicates more image detail and higher sharpness.

[0096] The objective quality evaluation results are shown in Table 1. Among the evaluation indicators, the highest values ​​of information entropy and average gradient are displayed in bold, and the next highest values ​​are displayed skewed. The lowest values ​​of natural image quality evaluation are displayed in bold, and the next lowest values ​​are displayed skewed.

[0097] Table 1

[0098]

[0099]

[0100] As shown in Table 1, all four image enhancement methods effectively enhanced the overall illumination of low-light images from underground coal mines. The LR3M image enhancement method, with its high Entropy and AG scores, demonstrates its ability to effectively enhance the brightness of low-light areas and recover detail information. However, its high NIQE score indicates a degree of distortion. The BioEnh image enhancement method is the only one among the four to achieve the highest Entropy and NIQE scores. This method enhances the overall brightness of the image and recovers texture and edge information in low-light areas, but it can lead to overexposure and distortion in some areas. The SRLLIE image enhancement method, with its superior Entropy and AG scores, fully demonstrates its effectiveness in improving image brightness and recovering detail information, resulting in high image clarity. However, it can also cause overexposure in near-light areas. The image enhancement method proposed in this invention has high entropy, the highest AG score and the lowest NIQE score. This means that the image enhancement method proposed in this invention can not only effectively improve the brightness of the image and suppress over-enhancement, but also recover the image detail information and improve the visual effect of the image. The above experiments have verified the effectiveness of the method proposed in this invention.

[0101] Example 2

[0102] This embodiment 2 describes a coal mine underground image enhancement system, which is based on the same inventive concept as the coal mine underground image enhancement method in embodiment 1.

[0103] Specifically, the underground coal mine image enhancement system includes the following modules:

[0104] The image deglow module is used to establish a glow imaging degradation model to obtain a glow image. It uses a layer separation method to remove the glow, avoids overexposure problems, and obtains a deglow image.

[0105] The improved SRLLIE module is used to improve the objective function of the low-light enhancement SRLLIE image enhancement method based on the deglow image, and iteratively solve the improved objective function to obtain an illumination map that suppresses overexposure and preserves structural details, and a denoised reflectance map.

[0106] The illuminance map optimization module is used to adjust the brightness of the illuminance map using the S-shaped gamma correction function to obtain an optimized illuminance map.

[0107] The image enhancement module is used to multiply the optimized illuminance map and reflectance map by a dot product according to Retinex theory to obtain the enhanced image.

[0108] It should be noted that the implementation process of the functions and roles of each functional module in the underground coal mine image enhancement system is detailed in the corresponding steps of the method in Example 1, and will not be repeated here.

[0109] Example 3

[0110] This embodiment 3 describes a computer device that includes a memory and one or more processors.

[0111] The memory stores executable code, which, when executed by the processor, is used to implement the steps of the coal mine underground image enhancement method in Embodiment 1 above.

[0112] In this embodiment, the computer device can be any device or apparatus with data processing capabilities, and will not be described in detail here.

[0113] Example 4

[0114] This embodiment 4 describes a computer-readable storage medium storing a program that, when executed by a processor, implements the steps of a method for enhancing images in underground coal mines.

[0115] The computer-readable storage medium can be an internal storage unit of any device or apparatus with data processing capabilities, such as a hard disk or memory, or an external storage device of any device with data processing capabilities, such as a plug-in hard disk, smart media card (SMC), SD card, flash card, etc.

[0116] Of course, the above description is only a preferred embodiment of the present invention. The present invention is not limited to the above-described embodiments. It should be noted that any equivalent substitutions or obvious modifications made by those skilled in the art under the guidance of this specification fall within the scope of this specification and should be protected by the present invention.

Claims

1. A method for enhancing images in underground coal mines, characterized in that, Includes the following steps: Step 1. Establish a glow imaging degradation model to obtain a glow image, use a layer separation method to remove the glow, suppress overexposure, and obtain a degloss image; Step 2. Based on the low-light enhancement SRLLIE method, the objective function of the low-light enhancement SRLLIE method is improved according to the deglow image. The improved objective function is solved iteratively to obtain the illumination map that suppresses overexposure and preserves structural details and the denoised reflection map. Step 3. Adjust the brightness of the illuminance map using the S-shaped gamma correction function to obtain the optimized illuminance map; Step 4. According to Retinex theory, multiply the optimized illuminance map by the reflectance map to obtain the enhanced image; Step 2 specifically involves: The improved objective function formula for the low-light enhancement SRLLIE method is as follows: (3) in, Represents a reflection diagram. Represents an illuminance diagram. Represents a noise graph. For the input image, Represents a glow image; , and For coefficients; and These represent the F-norm and L1 norm, respectively. It is a first-order differential operator. It is the adjusted gradient of the input image; In step 2, the objective function of the improved SRLLIE low-light enhancement method is solved iteratively by the alternating direction multiplier method. Introducing auxiliary variables Replace the second term of the objective function Then the objective function in formula (3) is transformed into the following form: (4) By introducing the Lagrange multiplier Z to eliminate the equality constraints, we obtain the augmented Lagrange function: (5) in, , for , for , Represents the matrix inner product. Indicates the scalar value of punishment.

2. The method for enhancing underground coal mine images according to claim 1, characterized in that, Step 1 specifically involves: A glow imaging degradation model is established, and glow images are obtained based on the glow imaging degradation model. The glow imaging degradation model is as follows: (1) in, This represents a degloss image; Glow image , As a light source, This represents the convolution operation. Represents the atmospheric point spread function; The glow discharge is removed using a layer separation method. The objective function of layer separation is: (2) in, This represents the objective function for performing layer separation. For first-order derivative filters in the horizontal and vertical directions It is a second-order Laplace filter; robust function ,in for , For fixed parameters; For smoothness weight parameters, This represents the r channel of the deglow image. This represents the g channel of the degloss image. This represents the b channel of the degloss image; Solve the objective function in the above formula (2) to obtain the degloss image.

3. The coal mine underground image enhancement method according to claim 2, characterized in that, In step 2, the objective function is solved iteratively step by step. The following is the calculation process of the k-th iteration solution for each sub-problem: Subproblem: Ignore the relationship between formula (5) and Irrelevant terms yield the following objective function: (6) in, This represents the result of the k-th iteration of L. Let N represent the result of the kth iteration; by restating the objective function in formula (6), it becomes a classic least squares problem: (7) in, , , , , , They are respectively , , , , , Vectorization, Let l represent a diagonal matrix with elements l. express The result of the kth iteration, This represents the result of the kth iteration of n; by analyzing formula (7) Differentiate, and divide the expression in formula (7) by... Other variables are treated as constants, and will affect Setting the derivative to 0, we get the following equation: (8) Where D is the discrete gradient operator. , This represents the result of the (k+1)th iteration of r; Subproblem: Ignore the relationship between formula (5) and Irrelevant terms yield the following objective function: (9) in, This represents the result of the (k+1)th iteration of R. This represents the result of the k-th iteration of Z. Let T represent the result of the k-th iteration; the solution of the (k+1)-th iteration of L is as follows: (10) in, for Vectorization, for Vectorization, express The result of the (k+1)th iteration, express The result of the kth iteration, This represents the result of the k-th iteration of t. This represents the result of the k-th iteration of z; Subproblem: Ignore the relationship between formula (5) and Irrelevant terms lead to the following objective function: (11) The solution for the (k+1)th iteration of N is: (12) Subproblem: Ignoring formula (5) and Irrelevant terms lead to the following objective function: (13) The solution for the (k+1)th iteration is: (14) in, , for The calculation is performed element by element; express The result of the kth iteration; Auxiliary matrix and penalty scalar Update via the following methods: (15) in, It is constant and Only when and The difference between or and The entire iteration will stop when the difference between the two values ​​is less than a preset threshold or when the maximum number of iterations is reached. After the iteration, an illumination map that suppresses overexposure and preserves structural details and a denoised reflectance map are obtained.

4. The coal mine underground image enhancement method according to claim 3, characterized in that, In step 3, the brightness of the illuminance map is adjusted using the S-type gamma correction function: (16) in, To optimize the illuminance map, γ is the luminance adjustment factor.

5. The coal mine underground image enhancement method according to claim 4, characterized in that, Step 4 specifically involves: According to Retinex theory, the enhanced image is obtained by multiplying the optimized illuminance map by the reflectance map: (17) in, This indicates an enhanced image.

6. A coal mine underground image enhancement system for implementing the coal mine underground image enhancement method as described in claim 1, characterized in that, The underground coal mine image enhancement system includes: The image deglow module is used to establish a glow imaging degradation model to obtain a glow image. It uses a layer separation method to remove the glow, avoids overexposure problems, and obtains a deglow image. The improved SRLLIE module is used to improve the objective function of the low-light enhancement SRLLIE method based on the deglow image, and iteratively solve the improved objective function to obtain an illumination map that suppresses overexposure and preserves structural details, and a denoised reflectance map. The illuminance map optimization module is used to adjust the brightness of the illuminance map using the S-shaped gamma correction function to obtain an optimized illuminance map. The image enhancement module is used to multiply the optimized illuminance map and reflectance map by a dot product according to Retinex theory to obtain the enhanced image.

7. A computer device comprising a memory and one or more processors, wherein the memory stores executable code, characterized in that, When the processor executes the executable code, it implements the steps of the coal mine underground image enhancement method as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the coal mine underground image enhancement method as described in any one of claims 1 to 5.

Images