Underwater image enhancement method based on color correction and brightness channel optimization and related device

Abstract

The application belongs to the technical field of image processing, and aims at the technical problems of insufficient color correction and excessive or insufficient contrast enhancement in the current underwater image quality enhancement method, and provides an underwater image enhancement method based on color correction and luminance channel optimization and related devices. First, the original underwater image is color corrected to effectively improve the color cast problem; the underwater image is converted from the RGB color space to the HSV space, the luminance channel V is decomposed into a base layer and a detail layer, the base layer is denoised, and the edge detail of the detail layer is enhanced to obtain an enhanced luminance channel; the histogram of the enhanced luminance channel is stretched to enhance the global contrast; then, the contrast-enhanced luminance channel, the hue channel H and the saturation channel S are combined, and the combined image is converted to the RGB space to obtain the final enhanced underwater image.

Description

Technical Field

[0001] This application belongs to the field of image processing technology, and relates to an underwater image enhancement method and related apparatus based on color correction and luminance channel optimization. Background Technology

[0002] Underwater imagery is a crucial method for underwater exploration, as clear underwater images can provide a wealth of valuable information, aiding in the perception of the underwater environment. However, due to the selective absorption of light by water and the scattering effects of suspended particles and plankton, natural light undergoes significant attenuation as it propagates through water. This leads to a decline in the quality of underwater images acquired by imaging systems, resulting in reduced contrast, dim image brightness, blurred details, and color distortion. These problems severely impact the ability to extract valuable features from underwater images, posing a significant challenge to underwater exploration. Therefore, researching effective underwater imaging image restoration techniques is of great practical importance.

[0003] Currently, methods for improving underwater image quality mainly focus on color correction and contrast enhancement. Commonly used white balance methods such as Grey-World, Max RGB, Shades of Grey, and Grey-Edge do not consider the attenuation differences of different wavelengths of light underwater, relying instead on assumptions about the light source color. When the image color characteristics do not conform to these assumptions, color restoration often fails. Existing contrast enhancement methods mainly include physically based methods, direct enhancement methods, and deep learning-based methods. Physically based methods consider the physical processes of image degradation, so the restored image depends on the estimation of model parameters, making the process complex. Deep learning-based methods have limited applicability due to a lack of datasets. In contrast, direct enhancement methods are simple and effective, but can lead to over-enhancement or under-enhancement of the image. Summary of the Invention

[0004] This application addresses the technical problems of current methods for improving underwater image quality, such as color correction methods being prone to color restoration failures, contrast enhancement methods being complex and having limited applicability, and over-enhancing or under-enhancing. It provides an underwater image enhancement method, system, device, and storage medium. An adaptive color balance algorithm is used to perform color correction on the original underwater image, improving the visual visibility of the underwater image by enhancing details and stretching contrast in the luminance channels of the HSV color space.

[0005] To achieve the above objectives, this application adopts the following technical solution:

[0006] In a first aspect, this application proposes an underwater image enhancement method based on color correction and luminance channel optimization, comprising:

[0007] An adaptive color balance algorithm is used to correct the color of the original underwater image, resulting in a color-corrected underwater image.

[0008] The color-corrected underwater image was converted from the RGB color space to the HSV color space.

[0009] In the HSV space, the luminance channel V is decomposed into a base layer and a detail layer;

[0010] The base layer is denoised, and the detail layer is edge-detail enhancement to obtain a detail-enhanced luminance channel;

[0011] The brightness channel with enhanced detail is histogram stretched and normalized to the desired stretching range to obtain the brightness channel with enhanced contrast.

[0012] The contrast-enhanced luminance channel, hue channel H, and saturation channel S are merged to obtain the merged image;

[0013] The merged image is converted to RGB space to obtain an enhanced underwater image.

[0014] Furthermore, the step of using an adaptive color balance algorithm to perform color correction on the original underwater image includes:

[0015] S1.1, Determine the main color channel of the original underwater image by comparing the average value of each color channel in the RGB color model;

[0016] S1.2, based on the attenuation difference between the main color channel and other color channels, compensate for the color channels whose attenuation exceeds the preset range to obtain the color-corrected underwater image.

[0017] Furthermore, before decomposing the luminance channel V into a base layer and a detail layer, the process also includes preprocessing the luminance channel V:

[0018] Convert the luminance channel V to the logarithmic domain and normalize it to the (0,1) range:

[0019] V log =log(V+0.001)

[0020]

[0021] Among them, V log V is the value after logarithmic transformation of the V channel. n V represents the normalized luminance channel. min and V max V log The maximum and minimum values.

[0022] Furthermore, the decomposition of the luminance channel V into a base layer and a detail layer includes:

[0023] Using a hybrid The decomposition model breaks down the luminance channel V into a base layer and a detail layer;

[0024] The mixture Decomposition model, including:

[0025]

[0026] V D =V n -V B

[0027] Among them, V n V represents the luminance channel normalized to the (0,1) range. B The base layer represents the luminance channel; λ1 represents a coefficient controlling the smoothness of the base layer; λ2 represents a coefficient controlling the smoothness of the detail layer; V D This represents the detail layer of the luminance channel; 1 indicates a unit vector where all elements are 1. Represents the gradient operator, This indicates an indicator function.

[0028] Further, the denoising of the base layer includes:

[0029] The base layer is denoised using bilateral filtering;

[0030] The bilateral filtering includes:

[0031]

[0032] Among them, V B ' represents the base layer after bilateral filtering, V B The base layer represents the input, w represents the filter weights, (i,j) represents the position of the pixel, and Ω(x,y) represents the neighborhood centered on the pixel (i,j).

[0033] w(i,j)=w s (i,j)w r (i,j)

[0034] Among them, w s (i,j) represents the spatial distance weight, w r (i,j) represents the pixel difference weight.

[0035] Furthermore, the edge detail enhancement of the detail layer includes:

[0036] The gradient image is extracted by convolving the Sobel gradient operators in four directions with the detail layer respectively.

[0037] The gradient images in four directions are fused to obtain the texture edge image of the detail layer, and then normalized.

[0038] The normalized texture edge image is overlaid with a linear image to obtain a brightness channel with enhanced details.

[0039] Further, stretching the histogram of the enhanced luminance channel includes:

[0040] Gaussian weights were designed based on the mean and standard deviation of the enhanced luminance channel;

[0041] The enhanced luminance channel is stretched using designed Gaussian weights;

[0042] The stretched luminance channel is cropped to retain the range of 0.5% to 99.5% of the histogram grayscale levels.

[0043] Secondly, this application proposes an underwater image enhancement system based on color correction and luminance channel optimization, comprising:

[0044] The color correction module is used to perform color correction on the original underwater image using an adaptive color balance algorithm to obtain a color-corrected underwater image.

[0045] The first conversion module is used to convert the color-corrected underwater image from the RGB color space to the HSV color space;

[0046] The decomposition module is used to decompose the luminance channel V into a base layer and a detail layer in HSV space;

[0047] An enhancement module is used to denoise the base layer and enhance the edge details of the detail layer to obtain a detail-enhanced luminance channel.

[0048] The stretching module is used to stretch the histogram of the detail-enhanced luminance channel and normalize it to the desired stretching range to obtain the contrast-enhanced luminance channel.

[0049] The merging module is used to merge the contrast-enhanced luminance channel, hue channel H, and saturation channel S to obtain the merged image;

[0050] The second conversion module is used to convert the merged image to RGB space to obtain an enhanced underwater image.

[0051] Thirdly, this application proposes an electronic device, comprising:

[0052] Memory, used to store computer programs;

[0053] A processor is used to implement the steps of the underwater image enhancement method based on color correction and luminance channel optimization when executing the computer program.

[0054] Fourthly, this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the underwater image enhancement method based on color correction and luminance channel optimization described above.

[0055] Compared with the prior art, this application has the following beneficial effects:

[0056] This application proposes an underwater image enhancement method based on color correction and luminance channel optimization. First, adaptive color correction is performed on the original underwater image to effectively improve color cast. Noise reduction is applied to the base layer, and edge detail enhancement is performed on the detail layer to improve the visibility of details in the underwater image. Furthermore, histogram stretching effectively enhances the contrast of the underwater image. This application can effectively improve the blue and green color cast in underwater images based on the attenuation difference between the primary color channel and other color channels. In addition, combining detail preservation and histogram stretching effectively enhances image details and contrast, improving the visibility of underwater images.

[0057] This application also proposes an underwater image enhancement system, electronic device, and computer-readable storage medium based on color correction and luminance channel optimization, which possesses all the advantages of the above-mentioned underwater image enhancement methods and facilitates the application and promotion of the underwater image enhancement method of this application. Attached Figure Description

[0058] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0059] Figure 1 This is a schematic diagram of the first process of the underwater image enhancement method based on color correction and luminance channel optimization in this application.

[0060] Figure 2 This is a schematic diagram of the second process of the underwater image enhancement method based on color correction and luminance channel optimization in this application.

[0061] Figure 3 This is a flowchart illustrating the processing of the original image in an embodiment of the underwater image enhancement method based on color correction and luminance channel optimization in this application.

[0062] Figure 4 Comparison images of different original underwater images before and after color correction in the embodiments of this application;

[0063] Figure 5 This is a comparison image of the brightness channel before and after enhancement in the embodiments of this application;

[0064] Figure 6 This is a comparison image of the enhanced underwater image and the original underwater image in the embodiments of this application;

[0065] Figure 7 This is a schematic diagram of the connection of the underwater image enhancement system based on color correction and luminance channel optimization in this application. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0067] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0068] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0069] In the description of the embodiments of this application, it should be noted that if terms such as "upper," "lower," "horizontal," or "inner" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use, they are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. In addition, terms such as "first" and "second" are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0070] Furthermore, the use of the term "horizontal" does not imply that the component must be absolutely horizontal, but rather that it can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0071] In the description of the embodiments of this application, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0072] When light propagates underwater, it is absorbed by water molecules and scattered by suspended particles, often resulting in color distortion, decreased contrast, and loss of detail in underwater images. To improve underwater image quality, this application proposes an underwater image enhancement method and related apparatus based on color correction and luminance channel optimization. The following detailed description, in conjunction with embodiments and accompanying drawings, further illustrates this application:

[0073] To address the aforementioned problems, this application proposes an underwater image enhancement method and related apparatus based on color correction and luminance channel optimization. The following detailed description, in conjunction with embodiments and accompanying drawings, further illustrates this application:

[0074] See Figure 1 This is a schematic diagram of the first process of an underwater image enhancement method based on color correction and luminance channel optimization according to this application, which may include:

[0075] S101, an adaptive color balance algorithm is used to perform color correction on the original underwater image to obtain a color-corrected underwater image.

[0076] Adaptive color balance algorithms are methods used to adjust the color balance of images. They automatically adjust colors based on the content and scene of the image to achieve better visual effects. This method typically relies on local or global features of the image, analyzing color distribution and differences to automatically adjust color parameters for better color balance.

[0077] S102 converts the color-corrected underwater image from the RGB color space to the HSV color space.

[0078] RGB and HSV color spaces are two commonly used color spaces with different characteristics and applications. The RGB color space describes colors based on combinations of the three primary colors: red, green, and blue, while the HSV color space describes colors based on three parameters: hue, saturation, and lightness. Converting from RGB to HSV color space requires a series of calculations and conversions, but can also be done directly using existing functions or methods.

[0079] S103, in HSV space, decomposes the luminance channel V into a base layer and a detail layer.

[0080] In practical applications, the base layer typically represents the macroscopic structure and main features of an image, while the detail layer represents the microscopic details and texture information. By decomposing the luminance channel into base and detail layers, images can be processed and analyzed more flexibly.

[0081] S104 performs noise reduction on the base layer and edge detail enhancement on the detail layer to obtain a detail-enhanced luminance channel.

[0082] Since the base layer typically contains the main structure and features of an image, it can be denoised using various denoising algorithms. The specific denoising process can be tailored to the nature of the noise and the content of the image by selecting appropriate parameters to remove noise while preserving image details.

[0083] Since the detail layer contains microscopic details and texture information of the image, it can enhance the image's sharpness and visual effect by improving its edge details, thus making the image clearer and more detailed.

[0084] S105 stretches the histogram of the detail-enhanced luminance channel and normalizes it to the desired stretching range to obtain the contrast-enhanced luminance channel.

[0085] A histogram of the luminance channel is a graph that represents the luminance distribution of an image. The horizontal axis represents the luminance level, and the vertical axis represents the number of pixels in the image that represent that luminance level. Through the histogram, one can intuitively understand the luminance and contrast distribution of an image.

[0086] S106, merge the contrast-enhanced luminance channel, hue channel H, and saturation channel S to obtain the merged image.

[0087] S107 converts the merged image to RGB space to obtain an enhanced underwater image.

[0088] To address the quality degradation problem of underwater images, this application proposes an underwater image enhancement method based on color correction and luminance channel optimization. First, the underwater image undergoes color correction to improve color cast. The color-corrected underwater image is then converted to the HSV color space, and the luminance channel V is decomposed into a base layer and a detail layer. Noise is denoised in the base layer, and edge detail enhancement is performed on the detail layer image. Histogram stretching is then applied to the edge-enhanced luminance channel to improve global contrast. Finally, the channels are merged to obtain the enhanced underwater image. This application effectively improves the blue and green color cast in underwater images. Furthermore, it effectively enhances image detail and contrast, improving the visibility of underwater images.

[0089] See Figure 2This is a schematic diagram of the second process of an underwater image enhancement method based on color correction and luminance channel optimization according to this application, which may include:

[0090] S201, an adaptive color balance algorithm is used to perform color correction on the original underwater image to obtain a color-balanced underwater image.

[0091] First, the dominant color channel of the underwater image is determined by comparing the average values ​​of each color channel in the RGB color model. The dominant color channel represents the color channel with the least attenuation. The average value of each color channel is defined as:

[0092]

[0093] Where M and N are the height and width of the original underwater image I, respectively, and λ represents the three color channels.

[0094] In underwater scenes, long-wavelength red light attenuates most severely, so it often appears blue or green. Therefore, the dominant color channel can be determined by comparing the average values ​​of the blue and green channels. The dominant color channel is defined as:

[0095]

[0096] in, Indicates the primary color channel. Indicates a green channel. This indicates the blue channel.

[0097] Adaptive compensation parameters are calculated based on the attenuation difference between the primary color channel and the channel with the most severe attenuation, thereby compensating for the color channel with the most severe attenuation to achieve ideal color balance. The adaptive color correction method designed in this application is as follows:

[0098]

[0099] Among them, I R I G and I B These represent the red, green, and blue channels of the original image, respectively. R ′、I B ′ and I G ′ represent the red, blue, and green channels after color correction, respectively, and |·| represents the absolute value operation.

[0100] like Figure 4 As shown, several original underwater images were compared before and after color correction, with (a) showing the image before color correction and (b) showing the image after color correction. The comparison results demonstrate that the color correction algorithm proposed in this application can effectively eliminate color cast in underwater images.

[0101] S202 converts the color-corrected underwater image from the RGB color space to the HSV color space, preprocesses the luminance channel V, and then uses a blending process. The decomposition model decomposes the preprocessed luminance channel V into a base layer and a detail layer.

[0102] The luminance channel V undergoes preprocessing, which includes logarithmic transformation and normalization. The purpose of this preprocessing is to adjust the dynamic range to a suitable interval to display more details. The process is as follows:

[0103] V log =log(V+0.0001) (4)

[0104]

[0105] Among them, V n V represents the normalized luminance channel. min and V max V log The maximum and minimum values.

[0106] Using a hybrid The decomposition model will use the V obtained in the above steps n Decompose into base layer and detail layer, and blend The decomposition model is as follows:

[0107]

[0108] Among them, V B V represents the base layer and the detail layer. D =V n -V B In equation (7), 1 represents a unit vector whose elements are all 1. gradient operator This indicates an indicator function that outputs a binary vector, where λ1 represents the coefficient controlling the smoothness of the base layer and λ2 represents the coefficient controlling the smoothness of the detail layer.

[0109] S203 uses bilateral filtering to denoise the base layer image, while using the Sobel directional gradient operator to enhance the edge details of the detail layer image.

[0110] Bilateral filtering has excellent edge preservation and noise reduction capabilities. The filtering process is as follows:

[0111]

[0112] Among them, V B Basic layer, V B'This is the base layer after bilateral filtering, where (i,j) represents the position of a pixel, Ω(x,y) is the neighborhood centered at pixel (i,j), and w is the filtering weight. The filtering weight w can be expressed as:

[0113] w(i,j)=w s (i,j)w r (i,j) (8)

[0114]

[0115]

[0116] Among them, w s (i,j) represents the spatial distance weight, w r (i,j) represents the pixel difference weight, σ s σ represents the standard deviation of the spatial domain. r This represents the standard deviation of the grayscale range.

[0117] Enhance edge details in the detail layer image:

[0118] Constructing multi-directional gradient operators, the Sobel directional gradient operator includes: horizontal gradient operator:

[0119]

[0120] Vertical gradient operator:

[0121]

[0122] Diagonal gradient operator:

[0123]

[0124] Anti-angle directional gradient template:

[0125]

[0126] Let G1, G2, G3, and G4 represent the gradient operators in the horizontal, vertical, diagonal, and anti-diagonal directions, respectively, and their effects on the detail layer V. D The extracted edge images are as follows:

[0127]

[0128] in, This represents a convolution operation. The gradient images from the four directions are fused to obtain the texture edge image of the detail layer.

[0129] E = E1 + E2 + E3 + E4 (12)

[0130] Then normalize E to the range of 0-1, using the same normalization method as in equation (6).

[0131] Finally, the normalized result E n With V D Linear superposition yields an enhanced detail layer V D ':

[0132] V D '=V D +E n (13)

[0133] like Figure 5 The image shown is a comparison of the luminance channel before and after enhancement. (a) is the image before luminance channel enhancement, and (b) is the image after luminance channel enhancement. Figure 5 As can be seen, the contrast of the enhanced luminance channel image is significantly improved, and the details are clearer than before enhancement, verifying the effectiveness of the contrast stretching and detail enhancement method proposed in this application.

[0134] S204, by constructing a Gaussian weighting function to stretch the histogram of the brightness channel after edge enhancement and normalizing it to the desired stretching range, enhances the contrast of the image.

[0135] (1) The noise-reducing base layer and the enhanced detail layer are merged to obtain the contrast-enhanced luminance channel V':

[0136] V' = V B +1.2V D (14)

[0137] (2) Design Gaussian weights based on the mean and standard deviation of the luminance channel V'. The Gaussian weights are defined as follows:

[0138]

[0139] Where (i,j) represents the position of the pixel. σ and σ' are the mean and standard deviation of V', respectively.

[0140] (3) Stretch V' using the obtained Gaussian weights:

[0141]

[0142] Where V1 is the value after stretching, and k is the enhancement factor that controls the degree of enhancement, which can be set to 1.

[0143] (4) Crop the histogram of V1 to retain the range of 0.5% to 99.5% of the gray levels of the histogram;

[0144] (5) Further stretch the cropped histogram to the 0-1 range:

[0145]

[0146] Among them, V 1min =0.5% * V1, V 1max =0.995% * V1, V 1min and V 1max These are the lower and upper bounds of the histogram after cropping, respectively. The luminance channel is used to enhance the final contrast.

[0147] S205 merges the contrast-enhanced luminance channel, the hue channel H in HSV space, and the saturation channel S in HSV space to obtain the merged image. Then, the merged image is converted from HSV space to RGB space to obtain the enhanced underwater image.

[0148] like Figure 3 The diagram shown is a flowchart of the underwater image enhancement method in this embodiment.

[0149] Figure 6 The image shown is a comparison of the enhanced underwater image and the original underwater image. (a) is the original underwater image, and (b) is the enhanced underwater image. Figure 6 As can be seen, in terms of visual effect, the enhanced underwater image processed by this application is more natural and the details are clearer, indicating that this application has a great advantage in solving problems such as color cast, reduced contrast and blurred details in underwater images.

[0150] like Figure 7 The diagram shown is a schematic of an underwater image enhancement system based on color correction and luminance channel optimization proposed in this application embodiment, which may include:

[0151] The color correction module is used to perform color correction on the original underwater image using an adaptive color balance algorithm to obtain a color-corrected underwater image.

[0152] The first conversion module is used to convert the color-corrected underwater image from the RGB color space to the HSV color space;

[0153] The decomposition module is used to decompose the luminance channel V into a base layer and a detail layer in HSV space;

[0154] An enhancement module is used to denoise the base layer and enhance the edge details of the detail layer to obtain an enhanced luminance channel.

[0155] The stretching module is used to stretch the histogram of the enhanced brightness channel and normalize it to the desired stretching range to obtain a contrast-enhanced brightness channel.

[0156] The merging module is used to merge the contrast-enhanced luminance channel, the hue channel H in HSV space, and the saturation channel S in HSV space to obtain the merged image.

[0157] The second conversion module is used to convert the merged image from HSV space to RGB space to obtain an enhanced underwater image.

[0158] It should be noted that, in the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of each module is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another device, or some features may be ignored or not executed. The modules described as separate components may or may not be physically separated. The components shown as modules may be one or more physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs.

[0159] Furthermore, in the various embodiments of the present invention, the modules can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit described above can be implemented in hardware or as a software functional unit.

[0160] An electronic device provided in this application includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the underwater image enhancement method based on color correction and luminance channel optimization as described in any of the above embodiments.

[0161] Another electronic device provided in this application embodiment may further include: an input port connected to a processor for transmitting multimodal data collected by an external acquisition device to the processor; a display unit connected to the processor for displaying the processor's processing results to the outside world; and a communication module connected to the processor for enabling communication between the electronic device and the outside world. The display unit may be a display panel, a laser scanner, or the like; the communication method used by the communication module includes, but is not limited to, Mobile High Definition Link (HML), Universal Serial Bus (USB), High Definition Multimedia Interface (HDMI), wireless connectivity: Wi-Fi, Bluetooth communication, Bluetooth Low Energy communication, and IEEE 802.11s-based communication technology.

[0162] This application provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the steps of the underwater image enhancement method based on color correction and luminance channel optimization as described in any of the above embodiments.

[0163] The computer-readable storage media involved in this application include random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs, or any other form of storage media known in the art.

[0164] For descriptions of relevant parts of the underwater image enhancement device, electronic device, and computer-readable storage medium based on color correction and luminance channel optimization provided in this application, please refer to the detailed descriptions of the corresponding parts in the underwater image enhancement method provided in this application, which will not be repeated here. Furthermore, parts of the technical solutions provided in this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0165] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An underwater image enhancement method based on color correction and luminance channel optimization, characterized in that, include: An adaptive color balance algorithm is used to perform color correction on the original underwater image to obtain the color-corrected underwater image: S1.

1. The main color channel of the original underwater image is determined by comparing the average value of each color channel in the RGB color model. S1.2, Based on the attenuation difference between the main color channel and other color channels, compensate for the color channels whose attenuation exceeds the preset range to obtain the color-corrected underwater image; The color-corrected underwater image was converted from the RGB color space to the HSV color space. In the HSV space, the luminance channel V is decomposed into a base layer and a detail layer; The base layer is denoised, and the detail layer is edge-detail enhancement to obtain a detail-enhanced luminance channel; The detail-enhanced luminance channel is histogram stretched and normalized to the desired stretching range to obtain a contrast-enhanced luminance channel. The histogram stretching of the detail-enhanced luminance channel includes: designing Gaussian weights based on the mean and standard deviation of the enhanced luminance channel; stretching the enhanced luminance channel using the designed Gaussian weights; and cropping the stretched luminance channel to retain the 0.5% to 99.5% range of histogram gray levels. The contrast-enhanced luminance channel, hue channel H, and saturation channel S are merged to obtain the merged image; The merged image is converted to RGB space to obtain an enhanced underwater image.

2. The underwater image enhancement method based on color correction and luminance channel optimization according to claim 1, characterized in that, Before decomposing the luminance channel V into a base layer and a detail layer, the process also includes preprocessing the luminance channel V: Convert the luminance channel V to the logarithmic domain and normalize it to the range (0, 1): in, This is the value after logarithmic transformation of the V channel. This represents the normalized luminance channel. and They represent The maximum and minimum values.

3. The underwater image enhancement method based on color correction and luminance channel optimization according to claim 2, characterized in that, The process of decomposing the luminance channel V into a base layer and a detail layer includes: Using a hybrid The decomposition model breaks down the luminance channel V into a base layer and a detail layer; The mixture Decomposition model, including: in, This represents the luminance channel normalized to the range (0, 1). The base layer representing the luminance channel. A coefficient representing the smoothness of the base layer. A coefficient representing the smoothness of the detail layer. The detail layer representing the luminance channel. This represents a unit vector whose elements are all 1s. Represents the gradient operator. This indicates an indicator function.

4. The underwater image enhancement method based on color correction and luminance channel optimization according to claim 3, characterized in that, The denoising of the base layer includes: The base layer is denoised using bilateral filtering; The bilateral filtering includes: in, This represents the base layer after bilateral filtering. The base layer representing the input. Indicates the filter weights. Indicates the position of a pixel. Represented by pixels The neighborhood centered on; in, Indicates spatial distance weights. This represents the pixel difference weight.

5. The underwater image enhancement method based on color correction and luminance channel optimization according to claim 4, characterized in that, The edge detail enhancement of the detail layer includes: Multi-directional Sobel gradient operators are used to convolve with detail layers to extract gradient images; Multi-directional gradient images are fused to obtain the texture edge image of the detail layer, and then normalized. The normalized texture edge image is linearly superimposed on the original detail layer to obtain a detail-enhanced luminance channel.

6. An underwater image enhancement system based on color correction and luminance channel optimization, characterized in that, include: The color correction module is used to perform color correction on the original underwater image using an adaptive color balance algorithm to obtain the color-corrected underwater image: S1.1, which determines the main color channel of the original underwater image by comparing the average value of each color channel in the RGB color model; S1.2, Based on the attenuation difference between the main color channel and other color channels, compensate for the color channels whose attenuation exceeds the preset range to obtain the color-corrected underwater image; The first conversion module is used to convert the color-corrected underwater image from the RGB color space to the HSV color space; The decomposition module is used to decompose the luminance channel V into a base layer and a detail layer in HSV space; An enhancement module is used to denoise the base layer and enhance the edge details of the detail layer to obtain a detail-enhanced luminance channel. The stretching module is used to perform histogram stretching on the detail-enhanced luminance channel and normalize it to the desired stretching range to obtain a contrast-enhanced luminance channel. The histogram stretching of the detail-enhanced luminance channel includes: designing Gaussian weights based on the mean and standard deviation of the enhanced luminance channel; stretching the enhanced luminance channel using the designed Gaussian weights; and cropping the stretched luminance channel to retain the 0.5% to 99.5% range of histogram gray levels. The merging module is used to merge the contrast-enhanced luminance channel, hue channel H, and saturation channel S to obtain the merged image; The second conversion module is used to convert the merged image to RGB space to obtain an enhanced underwater image.

7. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the underwater image enhancement method based on color correction and luminance channel optimization as described in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the underwater image enhancement method based on color correction and luminance channel optimization as described in any one of claims 1 to 5.

Images