Underwater image quality enhancement method based on coupling degradation processing

Abstract

The invention discloses an underwater image quality enhancement method based on coupling degradation processing. The method comprises the following steps: acquiring an underwater image data set; constructing a combined degradation processing model, and inputting the underwater image data set into the combined degradation processing model for training to obtain a trained combined degradation processing model; and obtaining a to-be-detected underwater image, and inputting the to-be-detected underwater image into the trained joint degradation processing model for underwater image enhancement to obtain a target underwater quality enhanced image. According to the method, the problem that the image quality cannot be effectively recovered when a traditional image processing method faces a complex coupling degradation problem is solved, efficient and accurate perception of the underwater robot to the environment is achieved, and therefore the operation safety of the underwater robot is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of underwater image enhancement, and in particular relates to an underwater image quality enhancement method based on coupled degradation processing. Background Art

[0002] Underwater image enhancement is a key technology in the field of image processing. Its effect directly affects the underwater robot's ability to perceive the environment, thereby affecting its operating safety. During underwater operations, the complexity and unpredictability of the marine medium environment, such as the absorption and scattering of light by water, the concentration of suspended particles, the presence of water flow and plankton, and the stirring of sediment during operations, will inevitably lead to various degradation phenomena in underwater images. These degradation phenomena do not exist independently, but are coupled with each other, significantly reducing the quality of underwater images. Although many existing methods perform well in specific degradation environments, in real scenes, due to the complexity of degradation characteristics and the existence of multiple couplings, a single method is difficult to achieve ideal restoration effects. In addition, the particularity of the underwater environment makes it difficult for manual intervention to effectively improve the imaging quality, which easily leads to the underwater robot's erroneous perception and abnormal judgment of the environment. This not only seriously affects the accuracy of underwater vision, but also has an adverse impact on subsequent image analysis and understanding tasks, and ultimately threatens the safety of underwater robot operation. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes an underwater image quality enhancement method based on coupled degradation processing, so as to realize efficient and accurate perception of the environment by the underwater robot, thereby improving the operation safety of the underwater robot.

[0004] To achieve the above object, the present invention provides an underwater image quality enhancement method based on coupled degradation processing, comprising: acquiring an underwater image data set;

[0005] Constructing a joint degradation processing model, inputting the underwater image data set into the joint degradation processing model for training, and obtaining a trained joint degradation processing model;

[0006] An underwater image to be tested is obtained, and the trained joint degradation processing model is input to perform underwater image enhancement to obtain a target underwater quality enhanced image.

[0007] Optionally, the joint degradation processing model includes: a joint feature extraction module, a probability guidance module, a feature post-processing module and a water area balance loss function;

[0008] The joint feature extraction module is used to mine the potential information of the underwater image and extract the coupled potential feature map through a multi-stage cascade design;

[0009] The probability guidance module is used to adjust the statistical properties of the coupled potential feature map, match the training and weight feature distributions, and output features;

[0010] The feature post-processing module is used to optimize the feature by adopting grayscale correction and boundary enhancement strategies;

[0011] The water area balance loss function is used to constrain the entire training process of the joint degradation processing model.

[0012] Optionally, the joint feature extraction module includes: a feature extraction unit, a first exponential linear unit-based compression and excitation residual unit, and an upsampling unit;

[0013] The feature extraction unit is used to perform nonlinear processing on the input underwater image to extract the initial underwater image features;

[0014] The first exponential linear unit-based compression and excitation residual unit is used to enhance the coupled latent feature representation through dynamic weight allocation and smooth nonlinear transformation according to the initial underwater image features;

[0015] The upsampling unit is used to restore the spatial scale of the coupled potential feature map.

[0016] Optionally, the probability guidance module includes: a probability generation unit, an adaptive instance normalization unit, and a second compression and excitation residual unit based on an exponential linear unit;

[0017] The probability generation unit is used to extract the required mean and standard deviation distribution from the coupled potential feature map after the scale is restored;

[0018] The adaptive instance normalization unit is used to adjust the feature map of the content image using the mean and variance of the target style image;

[0019] The second compression and excitation residual unit based on the exponential linear unit is used to optimize and dynamically adjust the required feature distribution.

[0020] Optionally, the feature post-processing module includes a grayscale correction component and a boundary enhancement mask strategy;

[0021] The grayscale correction component is used to correct the model's over-adjustment of the image, adjust the image's color balance, and eliminate unnatural color casts;

[0022] The boundary enhancement mask strategy is used to enhance detail information through Gaussian adjustment, reduce artifacts, and eliminate the loss caused by boundary deviation.

[0023] Optionally, the water area balance loss function is calculated as:

[0024] AnquBalanceLoss=|ABL out -ABL in +λ imp | 2

[0025] Among them, ABL out Indicates the value of the output image after relevant calculations during the training process; ABL in Represents the value of the input image after similar calculations during the training process; λ imp Indicates the amount of deviation.

[0026] Optionally, the feature extraction unit consists of a first convolutional layer, a first activation function, a random dropout layer and a pooling layer;

[0027] The first convolutional layer is used to extract local features of the input underwater image to obtain a primary feature map, the first activation function is used to introduce nonlinear characteristics, the random inactivation layer is used to prevent overfitting, the primary feature map is further downsampled through the pooling layer, and the initial underwater image features are extracted.

[0028] Optionally, the first exponential linear unit-based compression and excitation residual unit includes reflection padding, a second convolutional layer, a second activation function, an ESE layer and a residual connection, specifically including:

[0029] The first exponential linear unit-based compression and excitation residual unit uses reflection filling to mirror fill the edge of the feature map to ensure that the convolution operation does not introduce boundary artifacts;

[0030] The second convolutional layer and the second activation function are used to extract features, and the channel attention is adjusted through the ESE layer;

[0031] The ESE layer extracts global channel features by adaptive average pooling, and uses two of the second convolutional layers and the second activation function for nonlinear transformation;

[0032] Channel weighting is achieved through Sigmoid function;

[0033] The channel attention mechanism dynamically weights the feature contributions of different channels according to the global information of each channel, and further strengthens the feature expression through the residual connection.

[0034] Optionally, the upsampling unit includes a third convolutional layer, a third activation function and an upsampling layer;

[0035] The feature map is processed by the third convolution layer and the third activation function, and the spatial size of the feature map is increased by using the upsampling layer to restore the feature space resolution.

[0036] Technical effect of the invention: The invention discloses a method for enhancing underwater image quality based on coupled degradation processing, and uses a joint feature extraction module to mine potential information from various coupled degradations and capture the complexity of underwater images. The probability guidance module is used to adjust the overall statistical properties of underwater images to match the distribution of training features. The feature post-processing module is used to optimize and enhance the output features of the framework, reduce the information loss in the edge area, reduce the boundary deviation in the reasoning process, and enhance the clarity and details of the boundary while maintaining the overall image quality. The water balance loss is realized and the model is trained on the underwater public data set UIEB data based on the supervision of this loss function to ensure that the model can effectively learn the features of various coupled degradations and achieve a balance between color, clarity and contrast. The model performance is tested in different underwater scene data sets to check the enhancement effect of the model on different underwater scene images. So far, the task of enhancing the quality of underwater images in complex coupled degradation scenes has been achieved. The present invention can achieve effective processing of coupled degradation and has good generalization performance, which can help underwater robots achieve efficient and accurate perception of the environment, thereby improving their operating safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0038] Figure 1 A schematic diagram of a flow chart of an underwater image quality enhancement method based on coupled degradation processing according to an embodiment of the present invention;

[0039] Figure 2 This is a schematic diagram of the structure of a joint degradation processing model according to an embodiment of the present invention;

[0040] Figure 3 This is a schematic diagram of the structure of a joint feature extraction module according to an embodiment of the present invention;

[0041] Figure 4 This is a schematic diagram of the effectiveness visualization results of the joint degradation processing model according to an embodiment of the present invention on the U45 data set;

[0042] Figure 5 This is a schematic diagram of the visual comparison results of the joint degradation processing model of the present invention and other SOTA methods on the UIEB dataset. DETAILED DESCRIPTION

[0043] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0044] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0045] like Figure 1 As shown, this embodiment provides an underwater image quality enhancement method based on coupled degradation processing, including: acquiring an underwater image data set;

[0046] Construct a joint degradation processing model, input the underwater image data set into the joint degradation processing model for training, and obtain a trained joint degradation processing model;

[0047] The underwater image to be tested is obtained, and the trained joint degradation processing model is input to perform underwater image enhancement to obtain the target underwater quality enhanced image.

[0048] Furthermore, if Figure 2 The joint degradation processing model shown includes: a joint feature extraction module, a probability guidance module, a feature post-processing module and an aquatic balance loss function (AquaBalanceLoss);

[0049] A joint feature extraction module is used to mine the potential information of underwater images and extract coupled potential feature maps through a multi-stage cascade design;

[0050] The probability guidance module is used to adjust the statistical properties of the coupled latent feature map, match the training feature distribution, and output features;

[0051] Feature post-processing module, used to optimize features by using grayscale correction and boundary enhancement strategies;

[0052] AquaBalanceLoss is a function used to jointly handle the constraints of the entire training process of the degradation processing model.

[0053] Specifically, first, the input underwater image passes through a multi-layer feature extraction module to extract preliminary multi-scale features. After feature extraction, the features are dynamically weighted and enhanced through the compression and excitation residual unit based on the exponential linear unit (ESE-ResBlock). After joint feature extraction, the processed features are further subjected to probability estimation and distribution adjustment to ensure the stability of image quality, and the adaptability of style and content is enhanced through adaptive normalization. Finally, the image is post-processed through a grayscale correction block and a boundary enhancement mask to ensure that the grayscale and edge information of the image are optimized, reduce artifacts and enhance edge details, and finally generate high-quality underwater images.

[0054] First, the present invention introduces a multi-stage cascade design in the Joint Feature Extraction Module (JFEModule) to mine potential information from various coupled degradations. The encoding module takes into account the influence of factors such as color, illumination, and scattering to capture the complexity of underwater images. Afterwards, these feature maps are adjusted to match the distribution of training features through the Probabilistic Bootstrap Module (PB Module). To further optimize and enhance the output features of the framework, the present invention designs a Feature Post-Processing Module (FPP Module), which adopts grayscale correction and border enhancement mask (Border Enhancement Mask, BEM) strategies to reduce information loss in edge areas, solve the problem of boundary deviation during reasoning, and enhance the clarity and details of boundaries while maintaining overall image quality. Finally, in order to balance color, clarity, and contrast, and to ensure that the model can effectively learn the features of various coupled degradations, the present invention proposes a new water balance loss function (AquaBalanceLoss). This loss function supervises the learning process of the model and enhances its feature reasoning ability.

[0055] Furthermore, if Figure 3 The joint feature extraction module shown includes a feature extraction module, a first exponential linear unit based compression and excitation residual unit (ESE-ResBlock) and an upsampling unit (UpBlock);

[0056] A feature extraction module, used for introducing nonlinearity into the input underwater image to extract initial underwater image features;

[0057] The first one is an exponential linear unit based compression and excitation residual unit (ESE-ResBlock), which is used to extract the coupled latent feature map according to the initial underwater image features through dynamic weight allocation and smooth nonlinear transformation;

[0058] Upsampling unit, used to restore the scale of the coupled latent feature map.

[0059] Furthermore, the feature extraction unit includes a first convolutional layer (Conv), a first activation function (ELU), a random dropout layer (Dropout) and a maximum pooling layer (MaxPooling).

[0060] The input underwater image is first subjected to feature extraction through the first convolutional layer to obtain a preliminary feature representation. Then, nonlinear characteristics are introduced through the first activation function, so that the model can capture complex nonlinear relationships. The random dropout layer (Dropout) prevents overfitting and enhances the generalization ability of the model when processing complex images. Further feature representation is extracted through the second convolution and the first activation function. Finally, the maximum pooling layer (MaxPooling) reduces the spatial size of the feature map while retaining the most representative features for subsequent processing.

[0061] Furthermore, the first exponential linear unit-based compression and excitation residual unit (ESE-ResBlock) includes a reflection pad (ReflectionPad), a second convolutional layer (Conv), a second activation function (ELU), an ESE layer and a residual connection. First, the first exponential linear unit-based compression and excitation residual unit (ESE-ResBlock) structure uses reflection padding to mirror the edges of the feature map to ensure that the convolution operation does not introduce boundary artifacts. Then, the second convolutional layer and the second activation function (ELU) are used to extract features, and channel attention adjustment is performed through the ESE layer. The ESE layer extracts global channel features through adaptive average pooling, and uses two second convolutional layers and the second activation function (ELU) for nonlinear transformation, and finally realizes channel weighting through the Sigmoid function. The channel attention mechanism dynamically weights the feature contributions of different channels according to the global information of each channel, and further strengthens the feature expression through residual connections.

[0062] Furthermore, the upsampling unit includes a third convolutional layer (Conv), a third activation function (ELU) and an upsampling layer (Upsampling).

[0063] The upsampling unit restores the spatial resolution of the feature map through upsampling operations. First, the feature map is processed through the third convolution layer and the third activation function (ELU), and the upsampling layer is used to increase the spatial size of the feature map and restore the feature spatial resolution to provide support for subsequent modules.

[0064] Specifically, the present invention proposes a joint feature extraction module, which adopts an encoder-decoder structure and is composed of a feature extraction module (Feature Extraction Module, FE Module) and a first exponential linear unit-based compression and excitation residual unit (ESE-ResBlock), aiming to extract potential features from an input image. Figure 3As shown in Figure 1, by introducing nonlinearity and overfitting prevention measures, the learning and generalization capabilities of the model are enhanced. This lays the foundation for the JFE module to effectively extract and abstract input data features. In order to more efficiently utilize model parameters, enhance the model's adaptive feature recalibration capabilities, and achieve the extraction of key coupled degradation features and effective suppression of irrelevant features, such as Figure 2 As shown, the present invention proposes a first compression and excitation residual unit (ESE-ResBlock) based on exponential linear units. This module improves the representation ability of the model by modeling channel dependencies. This design enhances the model's focus on key features by recalibrating the convolution response and suppressing irrelevant features, and guides the model to give priority to restoring important features of multiple coupling degradation. The first compression and excitation residual unit (ESE-ResBlock) based on exponential linear units enhances the model's modeling ability of nonlinear distortion of underwater images by introducing nonlinearity, making the capture of nonlinear features more effective. In addition, it will also further enhance the model's adaptability and learning ability to image features under various environmental conditions. Finally, the first compression and excitation residual unit (ESE-ResBlock) based on exponential linear units realizes dynamic weight allocation and smooth nonlinear transformation, improves the adaptability of the model, enables it to handle multiple coupling degradation and correct nonlinear phenomena in underwater images, and ultimately improves the generalization performance of the model in diverse underwater environments. The present invention combines the first exponential linear unit-based compression and excitation residual unit (ESE-ResBlock) with the FE module to ensure the smooth flow of gradients, alleviate the problem of gradient disappearance during training, and improve the learning speed and stability of the model.

[0065] Furthermore, if Figure 3 The probability guided module shown includes a probability generation unit (PG Module), an adaptive instance normalization unit (AdaIN) and a second exponential linear unit based compression and excitation residual unit (ESE-ResBlock);

[0066] A probability generation unit, used to extract the required mean and standard deviation distribution according to the latent feature map after the scale is restored;

[0067] Adaptive Instance Normalization (AdaIN) unit, which is used to construct the required Gaussian distribution based on the desired mean and standard deviation distribution;

[0068] The second exponential linear unit based compression and excitation residual unit (ESE-ResBlock) is used to realign the required Gaussian distribution.

[0069] Specifically, the probability-guided module adaptively optimizes the color, contrast and macrostructure of the image through style adjustment and feature recalibration, while retaining the scene structure of the original content. The present invention combines the second exponential linear unit-based compression and excitation residual unit (ESE-ResBlock) with the adaptive instance normalization (AdaIN) unit to enhance the flexibility of the model in style adjustment and content preservation in multiple coupled degradation environments.

[0070] In the probability guidance module proposed in this invention, in order to further optimize the input features, a Probability Generation Module (PG Module) is designed to construct the required mean and standard deviation distribution. The feature map output by the joint feature extraction module (JFE Module) is set as the data matrix F1, Where B, C, H, and W represent the batch size, number of channels, height, and width, respectively.

[0071] First, calculate the mean μ of each channel f from the obtained feature map data matrix F1 f and standard deviation σ f ,get:

[0072]

[0073] and

[0074]

[0075] Among them, f ij Represents the pixel value at a specific position of channel f.

[0076] Afterwards, the invention uses a 1×1 convolution kernel to extract features from the mean vector, and the result is:

[0077]

[0078] and

[0079]

[0080] a represents the feature extracted from the mean vector through the convolution layer, and m represents the feature extracted from the standard deviation vector through the convolution layer.

[0081] The present invention also uses a 1×1 convolution kernel to extract features from the standard deviation vector to obtain the mean and standard deviation

[0082] Next, the present invention extends the application of Gaussian distribution to the input optimization and adjustment of the adaptive instance normalization (AdaIN) unit. The present invention uses the extracted mean a and standard deviation m parameters to construct an n-dimensional Gaussian distribution It is expressed as:

[0083]

[0084] Used to adjust the mean of the data.

[0085] Similarly, the present invention uses b and n to construct another n-dimensional Gaussian distribution Expressed as

[0086]

[0087] The standard deviation used to adjust the data.

[0088] Finally, the present invention is from N mean and N std The mean and standard deviation are extracted from the Gaussian distribution and reused as the optimized unique mean and standard deviation to form the input mean μ of AdaIN optimal :

[0089] μ optimal =ExtractMean(N mean ) (7)

[0090] With standard deviation σ optimal :

[0091] σ optimal =ExtractStd(N std ) (8)

[0092] Among them, ExtractMean() and ExtractStd() represent the operations of extracting the mean and standard deviation from the Gaussian distribution respectively.

[0093] After global adjustment, the integrated second exponential linear unit-based compression and excitation residual unit (ESE-ResBlock) continues to guide the model to focus on key features, enhance the consistency of the adaptive instance normalization (AdaIN) effect, and improve the adaptability of the model to different underwater degradation conditions.

[0094] Further, the feature post-processing module includes a grayscale correction component and a boundary enhancement mask strategy;

[0095] Grayscale correction component, used to correct the model's over-adjustment of the image;

[0096] Boundary enhancement mask strategy is used to enhance detail information through Gaussian adjustment and eliminate artifacts and boundary deviation losses.

[0097] Specifically, Figure 2 As shown in the figure, in order to reduce the information loss in the edge area, reduce the boundary deviation during the reasoning process, and enhance the texture details, a feature post-processing module (FPP Module) is invented and designed, which further enhances the feature representation based on the existing feature map. The module consists of a grayscale correction component and a boundary enhancement mask (BEM) strategy. The grayscale correction module aims to correct the problem of over-adjustment of the image by the first few layers of the network, while the boundary enhancement mask method enhances the detail information through Gaussian adjustment to solve the artifacts and boundary deviation loss problems encountered during the model reasoning process.

[0098] Suppose the feature map obtained by the probability generation unit is F2, which enters the feature post-processing module and is processed by white balance to obtain F3. A Gaussian filter is applied to F3 to extract its low-frequency component, which is recorded as F low :

[0099] F low =G(F3,ω) (9)

[0100] Wherein, G represents a Gaussian filter, and ω represents a balance parameter of the Gaussian filter.

[0101] The boundary enhancement mask is denoted as F BEM :

[0102] F BEM =F3-F low +λ (10)

[0103] Among them, λ is used to maintain the strength of the boundary enhancement mask BEM. Finally, the boundary enhancement mask BEM is combined with F3 to obtain the adjusted feature map F P :

[0104]

[0105] After convolution adjustment, the final module output is obtained as the final output image of the model, that is, the enhanced underwater image. At this point, the model reasoning process is completed.

[0106] Furthermore, the water balance loss function (AquaBalanceLoss) is calculated as:

[0107] AnquBalanceLoss=|ABL out -ABL in +λ imp | 2 (12)

[0108] Among them, ABL out Represents the value of the output image after relevant calculation, ABL inRepresents the value of the input image after similar calculations. imp Indicates the deviation. ABL consists of three parts: color index (L coi ), clarity index (L si ) and contrast index (L cti ). The basic expression of ABL is as follows:

[0109] ABL=c1×L coi +c2×L si +c3×L cti (13)

[0110] Where c1, c2 and c3 represent the weights assigned to each attribute, c1 = 0.029, c2 = 0.295 and c3 = 3.550.

[0111] For L coi , the formula is as follows:

[0112] L coi =-0.027×l+0.159×r (14)

[0113] Among them, the mean correlation l and variance correlation r can be expressed as:

[0114]

[0115] and

[0116]

[0117] Among them, μ α (·) represents the alpha trimmed mean of pixel values, s α (·) represents the variance of the pixel value after alpha clipping is applied, that is, the sum of squared deviations from the alpha clipping mean divided by the number of pixels. The difference between the first and second color components R G And the difference between the fourth and third color components Y B It can be obtained by the following formula:

[0118] R G =RG (17)

[0119] and

[0120]

[0121] Among them, R, G, and B represent the first color, second color, and third color channel values ​​of the image respectively.

[0122] For L si , which evaluates the clarity of an image by calculating the weighted sum of the edge maps obtained after applying Sobel edge detection to the image. The formula is as follows:

[0123]

[0124] EME i Represents the result of applying Sobel edge detection to a channel and computing the channel edge map (EME) strength. i represents the weighting coefficient of the channel. c represents the number of channels.

[0125] For L cti , which evaluates the contrast of underwater images by calculating the contrast within image blocks. The formula is:

[0126]

[0127] Among them, α is a parameter used to adjust the entropy value scale. Increasing α can enhance the randomness of the evaluation. Top is the difference between the maximum and minimum values ​​of each image block, indicating the contrast range of the image block. It is calculated as:

[0128] top=max-min (21)

[0129] bot is the sum of the maximum and minimum values ​​of each image block, which is used to normalize the contrast range. It is calculated as:

[0130] bot=max+min (22)

[0131] Among them, max represents the maximum value in each image block, and min represents the minimum value of the image block.

[0132] The contrast weight coefficient ω is defined as:

[0133]

[0134] It is used to perform weighted averaging of the contrast scores of each block. k1 and k2 are the number of blocks after the image is divided by the window in the height and width directions, respectively.

[0135] Among them, H loss Indicates the height of the tensor when calculating the loss, W loss Indicates the width of the tensor when calculating the loss, and win_size is the partition window size parameter.

[0136] Combined with other loss functions, the total loss function of the model is:

[0137] Loss = λ1 × loss vgg +λ2×loss kl +λ3×loss re +λ4×AnquBalanceLoss (24)

[0138] Among them, λ1, λ2, λ3, λ4 are weight coefficients, which are 0.025, 1, 0.1, 0.1, respectively, and loss vgg is the vgg loss function, loss kl is the divergence loss, loss re For reconstruction loss, AnquBalanceLoss is the water area balance loss function proposed in the present invention.

[0139] loss vgg The expression is:

[0140]

[0141] Among them, φ l (·) represents the feature map of the lth layer. N l It represents the size of the feature map of the lth layer, I(output) represents the output image generated by the model, I(target) represents the target image, and L is the total number of feature map layers.

[0142] loss kl The expression is:

[0143]

[0144] Among them, q(z|x) is the posterior distribution, which represents the distribution of the latent variable z under the condition of a given input x, and p(z) is the prior distribution, which is set to the standard normal distribution. represents the difference measure between q(z|x) and p(z), and ∫(·)dz represents the integration of the latent variable z.

[0145] loss re The expression is:

[0146]

[0147] like Figure 4 The figure below shows the effectiveness of the model in the coupled degradation scenario. After color correction, the model of the present invention effectively handles forward scattering and avoids a foggy appearance. Similarly, after color correction, the model in the figure below further enhances brightness and texture details.

[0148] like Figure 5As shown, the visual comparison results of the method of the present invention and other SOTA methods on the UIEB dataset are shown. The visual comparison of different methods includes HS2CM2A, Shallow-UWnet, PUIE (MC, MP), FiveAPlus, CWR, Spectroformer, NU2Net and the method of the present invention JDPNet from left to right. In the first and fifth rows, the method of the present invention not only corrects the color, but also successfully adjusts the brightness and eliminates the scattering effect. Other methods correct the color cast, but the scattering effect is insufficient. In the second and third rows, the method of the present invention achieves excellent descattering effect and performs well. The processed images show better clarity and saturation, surpassing GT. In the last row, after color correction, other methods introduce different degrees of distortion in biological details, but the method of the present invention effectively retains these details. In the sixth row, after enhancing the brightness, the method of the present invention eliminates the underwater scattering effect and corrects the second color hue, thereby obtaining a better visual distance. This verifies the effective processing ability of the model in an environment with multiple coupled degradations. In addition, the proposed method has good detail post-processing and shows excellent detail preservation capability in all images.

[0149] The comparison of JDPNet with other SOTA methods on the UIEB dataset is shown in Table 1. The present invention selects commonly used quantitative indicators to evaluate image quality. These indicators include peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), underwater image quality measure (UIQM) and underwater color image quality evaluation (UCIQE). PSNR and SSIM evaluate the overall image quality and visual similarity, while UIQM and UCIQE are specifically evaluated for the unique quality issues of underwater images.

[0150] Table 1

[0151]

[0152]

[0153] comprehensive Figure 5 From the qualitative visual comparison, it can be seen that the proposed method is the best performing model, demonstrating its superiority in different coupled degradation scenarios.

[0154] The method of the present invention shows the best performance on multiple datasets, including UIEB, EUVP, U45, C60 and UCCS. On the UIEB dataset, the method of the present invention achieved the highest performance in PSNR, SSIM, UIQM and UCIQE indicators, preliminarily proving the superiority of the method of the present invention. Due to the poor quality of some GT data, the method of the present invention does not strictly pursue consistency with the true value, so it fails to achieve overwhelming high scores in PSNR and SSIM. On the large-scale EUVP dataset, the method of the present invention achieved the highest scores in PSNR, UIQM and UCIQE indicators, and the second highest score in SSIM, demonstrating its strong generalization ability. On the other three datasets, the method of the present invention is significantly ahead in UIQM score, indicating an excellent balance between color, contrast and sharpness. In the highly challenging C60 dataset, the method of the present invention performs well, demonstrating its excellent ability to handle complex coupled degradation and difficult scenes. This success is attributed to the design of the overall framework of the present invention and the effectiveness of the novel loss function.

[0155] The present invention discloses an underwater image quality enhancement method based on coupled degradation processing, which uses a joint feature extraction module to mine potential information from various coupled degradations and capture the complexity of underwater images. A probability guidance module is used to adjust the overall statistical properties of underwater images to match the distribution of training features. A feature post-processing module is used to optimize and enhance the output features of the framework, reduce information loss in edge areas, reduce boundary deviations in the reasoning process, and enhance the clarity and details of boundaries while maintaining the overall image quality. A water balance loss is achieved and a model is trained on an underwater public data set UIEB data based on the supervision of this loss function to ensure that the model can effectively learn the features of various coupled degradations and achieve a balance between color, clarity and contrast. The model performance is tested in different underwater scene data sets to check the enhancement effect of the model on different underwater scene images. So far, the task of enhancing the quality of underwater images in complex coupled degradation scenes is achieved. The present invention can achieve effective processing of coupled degradation and has good generalization performance, and can help underwater robots achieve efficient and accurate perception of the environment, thereby improving their operating safety.

[0156] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for underwater image quality enhancement based on coupled degradation processing, characterized in that: include: Obtain underwater image dataset; Constructing a joint degradation processing model, inputting the underwater image data set into the joint degradation processing model for training, and obtaining a trained joint degradation processing model; An underwater image to be tested is obtained, and the trained joint degradation processing model is input to perform underwater image enhancement to obtain a target underwater quality enhanced image.

2. The underwater image quality enhancement method based on coupled degradation processing according to claim 1, characterized in that: The joint degradation processing model includes: a joint feature extraction module, a probability guidance module, a feature post-processing module and a water balance loss function; The joint feature extraction module is used to mine the potential information of the underwater image and extract the coupled potential feature map through a multi-stage cascade design; The probability guidance module is used to adjust the statistical properties of the coupled potential feature map, match the training and weight feature distributions, and output features; The feature post-processing module is used to optimize the feature by adopting grayscale correction and boundary enhancement strategies; The water area balance loss function is used to constrain the entire training process of the joint degradation processing model.

3. The underwater image quality enhancement method based on coupled degradation processing according to claim 2, characterized in that: The joint feature extraction module includes: a feature extraction unit, a first compression and excitation residual unit based on an exponential linear unit, and an upsampling unit; The feature extraction unit is used to perform nonlinear processing on the input underwater image to extract the initial underwater image features; The first exponential linear unit-based compression and excitation residual unit is used to enhance the coupled latent feature representation through dynamic weight allocation and smooth nonlinear transformation according to the initial underwater image features; The upsampling unit is used to restore the spatial scale of the coupled potential feature map.

4. The underwater image quality enhancement method based on coupled degradation processing according to claim 2, characterized in that: The probability guidance module includes: a probability generation unit, an adaptive instance normalization unit and a second compression and excitation residual unit based on an exponential linear unit; The probability generation unit is used to extract the required mean and standard deviation distribution from the coupled potential feature map after the scale is restored; The adaptive instance normalization unit is used to adjust the feature map of the content image using the mean and variance of the target style image; The second compression and excitation residual unit based on the exponential linear unit is used to optimize and dynamically adjust the required feature distribution.

5. The underwater image quality enhancement method based on coupled degradation processing according to claim 2, characterized in that: The feature post-processing module includes a grayscale correction component and a boundary enhancement mask strategy; The grayscale correction component is used to correct the model's over-adjustment of the image, adjust the image's color balance, and eliminate unnatural color casts; The boundary enhancement mask strategy is used to enhance detail information through Gaussian adjustment, reduce artifacts, and eliminate the loss caused by boundary deviation.

6. The underwater image quality enhancement method based on coupled degradation processing according to claim 2, characterized in that: The water balance loss function is calculated as: AnquBalanceLoss=|ABL out -ABL in +λ imp | 2 Among them, ABL out Indicates the value of the output image after relevant calculations during the training process; ABL in Represents the value of the input image after similar calculations during the training process; λ imp Indicates the amount of deviation.

7. The underwater image quality enhancement method based on coupled degradation processing according to claim 3, characterized in that: The feature extraction unit consists of a first convolutional layer, a first activation function, a random dropout layer and a pooling layer; The first convolutional layer is used to extract local features of the input underwater image to obtain a primary feature map, the first activation function is used to introduce nonlinear characteristics, the random inactivation layer is used to prevent overfitting, the primary feature map is further downsampled through the pooling layer, and the initial underwater image features are extracted.

8. The underwater image quality enhancement method based on coupled degradation processing according to claim 3, characterized in that: The first exponential linear unit-based compression and excitation residual unit includes reflection padding, a second convolutional layer, a second activation function, an ESE layer and a residual connection, specifically including: The first exponential linear unit-based compression and excitation residual unit uses reflection filling to mirror fill the edge of the feature map to ensure that the convolution operation does not introduce boundary artifacts; The second convolutional layer and the second activation function are used to extract features, and the channel attention is adjusted through the ESE layer; The ESE layer extracts global channel features by adaptive average pooling, and uses two of the second convolutional layers and the second activation function for nonlinear transformation; Channel weighting is achieved through Sigmoid function; The channel attention mechanism dynamically weights the feature contributions of different channels according to the global information of each channel, and further strengthens the feature expression through the residual connection.

9. The underwater image quality enhancement method based on coupled degradation processing according to claim 3, characterized in that: The upsampling unit includes a third convolutional layer, a third activation function and an upsampling layer; The feature map is processed by the third convolution layer and the third activation function, and the spatial size of the feature map is increased by using the upsampling layer. Restore feature space resolution.

Images