An enhanced visibility method for turbid underwater scenes

Abstract

This invention discloses an enhanced visualization method for de-turbidity removal in turbid underwater scenes. First, the turbidity factor of the water body is obtained as a quantitative basis for the degree of environmental turbidity. Then, the Turbidity Adaptive Contrast-Limited Histogram Equalization (TCLAHE) algorithm is applied to the original turbid image. The enhancement amplitude and clipping constraints of each sub-block are adjusted according to the turbidity factor, thereby effectively suppressing noise amplification under high turbidity conditions and fully enhancing local contrast under low turbidity conditions. Next, salient detail enhancement is employed to highlight important targets and texture information in the image. Finally, the enhanced image is fused with other preprocessed information at multiple scales, taking the turbidity factor into account in the fusion weight calculation, achieving both a high degree of de-turbidity removal and visual realism. This method is innovative compared to existing underwater de-fogging and enhancement techniques, achieving stable and clear enhancement effects in various turbid scenes, and is widely applicable to fields such as marine observation, underwater robot vision, and underwater artifact photography.

Description

Technical Field

[0001] This invention relates to an enhanced visualization method for removing turbidity from underwater scenes, belonging to the technical field of computer vision processing of underwater images. Background Technology

[0002] With the increasing demand for marine exploration and river monitoring, the quality of underwater images is particularly important for research and applications. Clear underwater images are fundamental for observing marine biological behavior, identifying species characteristics, and understanding seabed topography. Low-quality images can lead to misclassification of organisms or misjudgment of geological structures. However, the quality of underwater images is severely affected by the characteristics of the water medium. In turbid waters, due to the scattering and absorption of light by suspended particles, underwater images often exhibit a "haze" effect, characterized by low contrast, dull and bluish colors, blurred details, and a significant decrease in target discernibility. This degradation not only affects direct human observation but also reduces the performance of computer vision algorithms (such as underwater target detection and robot visual navigation).

[0003] To improve the visual quality of turbid underwater images, industry and academia have proposed various image deturbidity removal and enhancement techniques. For example, physics-based deturbidity removal methods utilize underwater imaging models and prior knowledge (such as dark channel priors and light propagation models) to estimate scene depth and medium transmittance, thereby restoring a clear image. However, such methods often require accurate estimation of multiple parameters (such as water attenuation coefficients and background light). When water turbidity varies or on-site calibration is difficult, model calculations may be inaccurate, leading to deviations in the restoration results. Image enhancement methods directly improve image quality through histogram equalization, white balance, and sharpening filtering. For instance, Contrast-Limited Adaptive Histogram Equalization (CLAHE) is effective in enhancing local contrast and is widely used in underwater image enhancement. However, the traditional CLAHE algorithm does not consider environmental turbidity differences, and in high-turbidity scenes with severe stray light, it may over-enhance noise or fail to adequately enhance details. Another approach is to fuse the original image with a color-corrected and contrast-enhanced version to balance natural color and sharpness. However, existing fusion algorithms typically use fixed weight calculation methods and lack awareness of water turbidity, thus failing to guarantee optimal results under different turbidity conditions.

[0004] Therefore, there is an urgent need for an enhanced visualization method for deturbidity removal in turbid underwater scenarios, which can adaptively adjust the deturbidity removal and enhancement strategies according to the turbidity of the water, fully removing the turbidity effect while avoiding over-processing. Summary of the Invention

[0005] The summary section of this application is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description section below. This summary section is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0006] To address the problems and shortcomings of existing technologies, this invention aims to provide an enhanced visualization method for deturbidity removal in turbid underwater scenes. The method involves obtaining a turbidity factor using a turbidimeter or image-based visual estimation, processing the original underwater image with an improved version of the standard CLAHE algorithm in conjunction with the turbidity factor to obtain a first intermediate image, then filtering out weak details from the edges and textures of the first intermediate image to obtain a second intermediate image, and finally fusing the second intermediate image and the original underwater image to reconstruct and output the enhanced underwater image. This invention achieves greater image enhancement gain in high turbidity conditions and retains more original information in low turbidity conditions, resulting in a natural and clear final result. This addresses the problems mentioned in the background section.

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

[0008] This invention discloses an enhanced visualization method for removing turbidity in turbid underwater scenes, comprising the following steps:

[0009] Step 1, in response to acquiring a turbid raw underwater image;

[0010] Step 2: Obtain the turbidity factor using a turbidity meter or by image visual feature estimation based on the original underwater image;

[0011] Step 3: The original underwater image is processed by turbidity adaptive contrast-limited histogram equalization based on the turbidity factor to obtain the first intermediate image.

[0012] Step 4: Extract high-frequency components and detect salient regions in the first intermediate image, and perform gain amplification on salient edges and textures based on the turbidity factor to obtain the second intermediate image;

[0013] Step 5: Select the original underwater image and the second intermediate image for multi-scale decomposition, and then fuse them according to the turbidity factor and image features at each scale.

[0014] Step 6: Reconstruct and output the fused result to obtain the final enhanced underwater image.

[0015] Preferably, the calculation of the turbidity factor in step 2 further includes the following steps:

[0016] Step 2.1: Scale the acquired original underwater image;

[0017] Step 2.2: Simultaneously extract the average gradient magnitude, standard deviation of image brightness, and average pixel saturation of the image in both the grayscale domain and the HSV color space.

[0018] Step 2.3: Calculate the turbidity score based on the average gradient magnitude of the image, the standard deviation of the image brightness, and the average pixel saturation of the image.

[0019] Step 2.4: Normalize the turbidity score to obtain the turbidity factor corresponding to the actual physical turbidity.

[0020] Preferably, step 3 further includes the following steps:

[0021] Step 3.1: Divide the original underwater image into several grid sub-blocks;

[0022] Step 3.2: Remove thresholds based on the calibration benchmark of the clean water sample. and reference enhancement gain coefficient ;

[0023] Step 3.3: Obtain the corrected pruning threshold based on the modulation function that is monotonically correlated with the turbidity factor. and enhanced gain coefficient ;

[0024] Step 3.4, based on the corrected pruning threshold and enhanced gain coefficient Pixel histogram of redistributed grid sub-blocks.

[0025] Preferably, step 4 further includes the following steps:

[0026] Step 4.1: Extract high-frequency details from the first intermediate image;

[0027] Step 4.2: Use the Laplace or Sobel operator to obtain the edge map of the first intermediate image;

[0028] Step 4.3: Extract the fine texture structure of the first intermediate image by high-pass filtering;

[0029] Step 4.4 introduces visual saliency detection or edge amplitude screening to identify weak textures and edges in edge maps and fine texture structures;

[0030] Step 4.5: Apply detail gain to the filtered edge map and fine texture structure and superimpose it onto the first intermediate image to obtain the second intermediate image.

[0031] Preferably, step 5 further includes the following steps:

[0032] Step 5.1: Perform multi-scale decomposition on the original underwater image and the second intermediate image respectively to obtain sub-band representations from low frequency to high frequency;

[0033] Step 5.2: Set the fusion weighting factor based on the turbidity factor;

[0034] Step 5.3: At each scale, fuse the corresponding sub-bands of the original underwater image and the second intermediate image according to the fusion weight factor.

[0035] Preferably, in step 3.3, the turbidity factor is... Substituting the monotonically correlated increasing function and decreasing function ,based on and Corrected pruning threshold and enhanced gain coefficient ;

[0036] When turbidity factor At that time, the threshold for pruning was removed. and enhanced gain coefficient Close to the baseline value;

[0037] When turbidity factor When the threshold is between 0.3 and 0.7, the pruning threshold is... With enhanced gain coefficient It exhibits linear contraction and expansion;

[0038] When turbidity factor At that time, the threshold for pruning was removed. Approximately 30% of the reference value, enhancing the gain coefficient. It has increased to nearly 1.6 times.

[0039] Preferably, in step 4.5, based on the turbidity factor Controlling the detail gain coefficient The detail gain coefficient Limited to the interval [0.4, 1.2], it is represented as ,in , ;

[0040] In turbidity factor When close to 0, the detail gain coefficient A value close to 1.2 is sufficient to highlight the texture;

[0041] In turbidity factor When it is close to 1, the detail gain coefficient The value was reduced to near 0.4 to suppress false edges caused by suspended particles.

[0042] Preferably, in step 5.2, based on the turbidity factor Determine the fusion weight factor , represented by linear modulation as ,in , ;

[0043] As a second aspect of this application, the present invention also discloses an enhanced visualization system for deturbidity removal in turbid underwater scenes, comprising:

[0044] The turbidity analysis module is used to obtain the turbidity factor from the acquired raw underwater images using a turbidity meter or by estimating the image visual features.

[0045] The turbidity adaptive contrast enhancement module is used to perform turbidity adaptive contrast limited histogram equalization on the original underwater image according to the turbidity factor to obtain a first intermediate image.

[0046] A significant detail enhancement module is used to extract high-frequency components and detect significant regions in the first intermediate image, and to perform gain amplification on significant edges and textures based on the turbidity factor to obtain a second intermediate image;

[0047] The multi-scale fusion module is used to perform multi-scale decomposition on the original underwater image and the second intermediate image, and to assign fusion weights at each scale according to the turbidity factor and image features, and then output the corresponding fused reconstruction.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0049] This invention provides an enhanced visualization method for deturbidity removal in turbid underwater scenes. It utilizes an improved turbidity-adaptive contrast-limited histogram equalization algorithm (TCLAHE) that adjusts the amplitude and threshold of histogram equalization in real-time based on water turbidity, overcoming the imbalance in performance of traditional fixed enhancement methods across different turbidity levels. This approach ensures contrast enhancement while avoiding excessive noise introduction. Combining salient detail enhancement and multi-scale fusion, the two-stage processing highlights details of important targets while maintaining overall image naturalness. The fusion strategy dynamically configures weights based on turbidity perception, avoiding the drawbacks of over-sharpening details or color distortion that may occur with simple enhancement methods. Furthermore, this invention integrates environmental turbidity information throughout the image enhancement process, particularly coupling the degree of deturbidity removal with the fusion weights during the fusion stage—a key innovation compared to existing fusion algorithms. This mechanism ensures that the output image achieves the optimal balance between clarity and realism regardless of whether the water is clear or turbid. Because of the introduction of turbidity factor, the method of the present invention can automatically adjust parameters according to the on-site water quality conditions, and is therefore applicable to different types of water bodies (clear seawater, turbid rivers, silty ponds, etc.), and has good robustness and practical value. Attached Figure Description

[0050] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application.

[0051] In the attached diagram:

[0052] Figure 1 This is a diagram showing the main steps of the enhanced visibility method for removing turbidity in an embodiment of the present invention.

[0053] Figure 2 This is an overall flowchart of the enhanced visibility method for removing turbidity in an embodiment of the present invention;

[0054] Figure 3 In the embodiments of the present invention, (a)-(d) are comparison diagrams of the enhancement effect before and after under medium turbidity conditions of NTU13-16;

[0055] Figure 4 In the embodiments of the present invention, (e)-(g) are comparison diagrams of the effects before and after enhancement under medium turbidity conditions of NTU17-19, and (h) is a comparison diagram of the effects before and after enhancement under extremely high turbidity conditions of NTU40.

[0056] Figure 5 In this embodiment of the invention, (a) is the input original underwater image, (b)-(k) are comparison diagrams of the visual effects before and after enhancement using different methods under different turbidities, and (l) is the output effect diagram of the method of the present invention. Detailed Implementation

[0057] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0058] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0059] This invention discloses an enhanced visualization method for removing turbidity in turbid underwater scenes. The following will describe this disclosure in detail with reference to the accompanying drawings and embodiments. (Refer to...) Figure 1 and Figure 2 As shown, the main steps include:

[0060] Step 1, in response to acquiring a turbid raw underwater image;

[0061] Step 2: Obtain the turbidity factor using a turbidity meter or by image visual feature estimation based on the original underwater image;

[0062] Step 3: The original underwater image is processed by turbidity adaptive contrast-limited histogram equalization based on the turbidity factor to obtain the first intermediate image.

[0063] Step 4: Extract high-frequency components and detect salient regions in the first intermediate image, and perform gain amplification on salient edges and textures based on turbidity factor to obtain the second intermediate image;

[0064] Step 5: Select the original underwater image and the second intermediate image for multi-scale decomposition, and then perform corresponding fusion weights based on the turbidity factor and image features at each scale.

[0065] Step 6: Reconstruct and output the fused result to obtain the final enhanced underwater image.

[0066] This invention discloses a method for responding to the acquisition of turbid raw underwater images, based on the acquired turbidity information of the water body, and then using this information to determine subsequent color correction steps. First, the turbidity factor is obtained through turbidity measurement or image analysis of the environmental water body. In this embodiment, the turbidity factor can be calculated using a turbidimeter and an image-based visual estimation method. When conditions permit, a turbidimeter can be used to directly measure the physical turbidity of the water body, obtaining a turbidity value expressed in Niflöh units (NTU), which can then be normalized to a turbidity factor usable by the algorithm. However, turbidimeters are relatively expensive to use, mainly due to their high-precision sensors and real-time monitoring capabilities. Therefore, in the absence of dedicated turbidimeter sensors, we can use the image-based visual estimation method disclosed in this invention to calculate the equivalent turbidimetric factor. .

[0067] The image visual estimation method in this invention determines water turbidity by analyzing the sharpness and color distribution of the original image, calculating statistical features such as edge gradient intensity, brightness standard deviation, and color saturation, and constructing a score reflecting the level of turbidity by combining these features. It also includes the following steps:

[0068] Step 2.1: Scale the acquired raw underwater image;

[0069] Step 2.2: Simultaneously extract the average gradient magnitude, standard deviation of image brightness, and average pixel saturation of the image in both the grayscale domain and the HSV color space.

[0070] Step 2.3: Calculate the turbidity score based on the average gradient magnitude of the image, the standard deviation of the image brightness, and the average pixel saturation of the image.

[0071] Step 2.4: Normalize the turbidity score to obtain the turbidity factor corresponding to the actual physical turbidity.

[0072] Specifically, the original underwater image obtained in step 1 is first scaled to a uniform size. Then, three statistical features are extracted from both the grayscale and HSV color spaces: the average gradient magnitude, the standard deviation of image brightness, and the average pixel saturation. The turbidity score is then expressed as follows:

[0073] ;

[0074] in, This represents the average gradient magnitude of the image, which reflects the sharpness of the image edges. The standard deviation of image brightness reflects the image's contrast. This represents the average pixel saturation of an image, reflecting the vibrancy of its colors. Represented as weighting coefficients, This represents a minimum constant to avoid division by zero. The average gradient magnitude of the image. It is obtained by applying the Sobel operator to the grayscale image to calculate the horizontal and vertical gradients, and then averaging the gradient magnitudes of each pixel. Image brightness standard deviation. The global standard deviation is then taken from the luminance channel in Lab or HSV space. The RGB image is converted to Lab or HSV space, and the channel data is extracted before calculating the global standard deviation. The average pixel saturation of the image is then considered. The average value of the S channel in the HSV color space is taken. The RGB image is converted to the HSV color space, and the S channel data is extracted to calculate the matrix and find the mean. The S channel represents the color purity.

[0075] Turbidity score Perform maximum-minimum range normalization to obtain the turbidity factor corresponding to the actual physical turbidity. Turbidity score A linear transformation to a specified range is represented as [0, 1]. This is equivalent to turbidity scoring. Convert to a turbidity factor within the interval [0, 1] to eliminate dimensional differences and make its value consistent with the turbidity factor. Corresponding in dimensions. When water turbidity is higher, images tend to have lower gradients and brightness contrast, and weaker color saturation; therefore, turbidity scoring... It will increase, while the turbidity score will increase in clear water. The turbidity factor obtained in this way is relatively small. This can be understood as a measure of the degree of image degradation caused by the aquatic environment, which can be used for subsequent steps.

[0076] This invention improves upon the contrast-limited histogram equalization (CLAHE) algorithm, proposing a turbidity-adaptive CLAHE algorithm to perform local contrast enhancement on the original underwater image. The CLAHE histogram equalization method divides the image into several small grid blocks, performs histogram equalization on each sub-block individually, and limits the maximum threshold of its histogram pixel frequency to avoid excessive noise enhancement. The improvement of the turbidity-adaptive CLAHE algorithm in this invention compared to the standard CLAHE algorithm lies in its application of a global turbidity factor. The histogram stretching and contrast limit parameters for each sub-block are dynamically adjusted. This includes the following steps:

[0077] Step 3.1: Divide the original underwater image into several grid sub-blocks;

[0078] Step 3.2: Remove thresholds based on the calibration benchmark of the clean water sample. and reference enhancement gain coefficient ;

[0079] Step 3.3: Obtain the corrected pruning threshold based on the modulation function that is monotonically correlated with the turbidity factor. and enhanced gain coefficient ;

[0080] Step 3.4, based on the corrected pruning threshold and enhanced gain coefficient Pixel histogram of redistributed grid sub-blocks.

[0081] Specifically, a baseline pruning threshold is set in advance. and reference enhancement gain coefficient We used a water sample calibration experiment to determine the baseline pruning threshold. and reference enhancement gain coefficient The default value is 1, corresponding to the standard contrast-limited histogram equalization method CLAHE. For each grid sub-block of the current raw underwater image, the normalized turbidity factor is... Use the modified pruning threshold and enhanced gain coefficient This is used to redistribute the pixel histogram of the sub-block. Pruning threshold. Represented as and the enhanced gain coefficient Represented as .in, and This is a modulation function that is monotonically correlated with the turbidity factor. Selected as a decreasing function Choosing an increasing function reduces the pruning threshold when the turbidity factor is large.

[0082] Furthermore, the normalized turbidity factor Substituting the increasing and decreasing functions, the linear modulation curve is expressed as follows: and ,based on and This allows us to obtain the corrected pruning threshold. . With turbidity factor exist Monotonically decreasing over the interval, enhancing the gain coefficient exist The interval monotonically increases. When the turbidity factor... At low turbidity, remove threshold and enhanced gain coefficient Close to the benchmark value and When turbidity factor When the turbidity is between 0.3 and 0.7 (medium turbidity), the pruning threshold is... With enhanced gain coefficient They exhibit linear contraction and amplification, respectively. When the turbidity factor... (At high turbidity) threshold removal Only about 30% of the baseline value, enhancing the gain coefficient. It was increased by nearly 1.6 times to compensate for the dim local contrast caused by smog.

[0083] Pruning threshold Used to truncate the pixel count of a sub-block histogram to prevent high turbidity factors. The noise in the scene is over-equalized. (This affects the gain coefficient.) This is then used as a power-law gain to amplify or suppress the overall equilibrium result. Turbidity factor The pruning threshold is scaled only by the modulation function described above. With enhanced gain coefficient No further calculations are required; these two parameters only need to be updated once before calling TCLAHE. After processing by the turbidity-adaptive contrast-limited histogram equalization method TCLAHE in step 3, the first intermediate image can be obtained. This significantly improves the local contrast of the original underwater image, brightens dark details, enriches bright areas, and initially reduces the hazy effect caused by turbidity.

[0084] Then, building on the increased contrast, we further enhanced the image's details and textures to highlight prominent targets in the underwater scene. This specifically includes the following steps:

[0085] Step 4.1: Extract high-frequency details from the first intermediate image;

[0086] Step 4.2: Use the Laplace or Sobel operator to obtain the edge map of the first intermediate image;

[0087] Step 4.3: Extract the fine texture structure of the first intermediate image using high-pass filtering;

[0088] Step 4.4 introduces visual saliency detection or edge amplitude screening to identify weak textures and edges in edge maps and fine texture structures;

[0089] Step 4.5: Apply detail gain to the filtered edge image and fine texture structure and overlay it onto the first intermediate image to obtain the second intermediate image.

[0090] First, high-frequency detail extraction is performed on the first intermediate image obtained after enhancement using the improved TCLAHE algorithm. The edge map is obtained using either the Laplace or Sobel operator. The Sobel operator synthesizes the edge map by calculating the gradients in the x and y directions, while the Laplace operator detects edges using second-order differentiation; the Laplace operator is more sensitive to fine edges. Next, a high-pass filter is used to extract fine texture structures. By allowing high-frequency signals to pass through and suppressing low-frequency components, the resulting high-frequency components represent the detail information in the image. Then, the magnitude of detail enhancement is determined based on saliency. A visual saliency detection model is introduced, or important edges and textures are selected simply based on edge amplitude, i.e., filtering out weak textures and edges caused by noise. Next, detail gain is applied to the selected edge map and fine texture structures; that is, the high-frequency components are amplified by a certain proportion and superimposed back onto the first intermediate image to obtain the second intermediate image. Let... This represents the filtered edge map and fine texture structure, while If the detail gain coefficient is used, then the update of the first intermediate image pixel value is expressed as: .

[0091] Furthermore, detail gain coefficient Based on turbidity factor Controlling the detailed gain coefficient It is always limited to the range [0.4, 1.2]. Specifically, 、 ,Right now In ultra-low turbidity factor Approaching 0 At that time, detail gain coefficient Approximately 1.2 To fully highlight the texture. And in ultra-high turbidity factors... Approaching 1 At that time, detail gain coefficient The value was reduced to approximately 0.4 to suppress false edges caused by suspended particles. This salient detail enhancement process makes the outlines of key targets and surface details in the image clearer, such as the outlines of underwater organisms and the texture of the base rock, thus laying the foundation for subsequent fusion.

[0092] To adaptively consider the impact of water turbidity on deturbidity requirements during image fusion, this invention introduces a turbidity-aware fusion strategy in the weight calculation. Its core is to couple the degree of image deturbidity with the fusion weights to ensure that the fusion result both maximizes the elimination of turbidity and maintains visual naturalness. Specifically, it includes the following steps:

[0093] Step 5.1: Perform multi-scale decomposition on the original underwater image and the second intermediate image to obtain sub-band representations from low frequency to high frequency;

[0094] Step 5.2: Set the fusion weighting factor based on the turbidity factor;

[0095] Step 5.3: At each scale, fuse the corresponding sub-bands of the original underwater image and the second intermediate image according to the fusion weight factor.

[0096] Specifically, the original underwater image and the detail-enhanced image obtained through the aforementioned enhancement process are fused at multiple scales to combine the advantages of both and output the final result. The fusion employs multi-scale image fusion methods such as Laplacian pyramid or wavelet decomposition, that is, the original underwater image and the second intermediate image are decomposed at multiple scales to obtain a series of sub-band representations from low frequency to high frequency. The high-frequency sub-bands contain detailed information such as edges and textures. A fusion weight factor is set. The fusion weighting factor Based on turbidity factor The process involves determining the fusion weights, then synthesizing corresponding sub-bands of the original underwater image and the second intermediate image at each scale, and finally reconstructing the fused image.

[0097] Furthermore, the fusion weighting factor Based on turbidity factor To determine. Linear modulation is used. ,in 、 Turbidity factor This serves as a global adjustment factor influencing the weighting ratio. During fusion, a Laplacian pyramid is constructed for the original underwater image and the second intermediate image, and the weights are adjusted at each scale according to the fusion weighting factor. Select enhanced image subband, press Select sub-bands of the original image, synthesize them, and reconstruct the output fused image, which is the enhanced underwater image.

[0098] The turbidity-sensing multi-scale fusion proposed in step 5 allows the enhanced processing results to fully play their role in scenes where deturbidity is most needed, while preserving the authenticity of the original image in clearer scenes. The fused image not only greatly eliminates the haze and reveals rich and clear details such as object edges and textures, but also maintains overall tone and brightness harmony without over-sharpening or artifacts.

[0099] Through the above steps, this invention achieves efficient image deturbidity removal and enhancement for turbid underwater scenes. The final enhanced underwater image, compared to the original turbid underwater image, exhibits higher contrast, more vivid colors, and clearer details, significantly improving the visibility of underwater scenes. This method can be applied to various fields such as underwater robot vision, submersible photography, and ocean exploration, helping to obtain more reliable image information in complex and turbid water environments. (Refer to...) Figures 3 to 5 As shown, Figure 3 (a)-(d) are comparison graphs showing the effect of enhancement before and after under medium turbidity conditions in NTU13-16. Figure 4 (e)-(g) are comparison graphs showing the effect of enhancement before and after under medium turbidity conditions in NTU17-19. Figure 4 (h) is a comparison of the effects of enhancement before and after under extremely high turbidity conditions of NTU40. Figure 5 (a) is the original underwater image as input. Figure 5 (b)-(k) are comparison diagrams of the visual effects before and after enhancement using different methods at different turbidities. Figure 5 (l) is a diagram showing the output effect of the method of the present invention.

[0100] To achieve the above embodiments, this invention proposes an enhanced visualization system for deturbidity removal in turbid underwater scenes, comprising: a turbidity analysis module, used to acquire the original underwater image and obtain the turbidity factor using a turbidity meter or through image visual feature estimation; a turbidity adaptive contrast enhancement module, used to perform turbidity adaptive contrast-limited histogram equalization on the original underwater image based on the turbidity factor to obtain a first intermediate image; a salient detail enhancement module, used to extract high-frequency components and detect salient regions in the first intermediate image, and perform gain amplification on salient edges and textures based on the turbidity factor to obtain a second intermediate image; and a multi-scale fusion module, used to perform multi-scale decomposition on the original underwater image and the second intermediate image, and assign fusion weights at each scale based on the turbidity factor and image features, and then output the corresponding fused reconstruction.

[0101] To implement the above embodiments, the present invention also proposes an electronic device. This electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage device into a random access memory (RAM). Various programs and data required for the operation of the electronic device are also stored in the RAM. The processing unit, ROM, and RAM are interconnected via a bus. An input / output (I / O) interface is also connected to the bus. Typically, the following devices can be connected to the I / O interface: input devices including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices including, for example, magnetic tapes, hard disks, etc.; and communication devices. The communication device allows the electronic device to communicate wirelessly or wiredly with other devices to exchange data.

[0102] Furthermore, the present invention also proposes a computer storage medium. The computer storage medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0103] In this invention, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, the computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. This propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer storage medium other than a computer-readable storage medium, capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer storage medium can be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof. The computer storage medium can be included in the aforementioned electronic device or can exist independently without being assembled into the electronic device. The computer storage medium carries one or more programs that, when executed by the electronic device, enable the electronic device to implement an enhanced visualization method for deturbidity removal in turbid underwater scenes.

[0104] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. An enhanced visibility method for dehazing a turbid underwater scene, characterized in that, The method comprises the following steps: Step 1, in response to obtaining a turbid original underwater image; Step 2, obtaining a turbidity factor by using a turbidimeter or image visual feature estimation processing based on the original underwater image; Step 3, performing turbidity self-adaptive contrast limited histogram equalization processing on the original underwater image according to the turbidity factor to obtain a first intermediate image; the specific implementation steps are as follows: Step 3.1, dividing the original underwater image into a plurality of grid sub-blocks; Step 3.

2. Calibrating the baseline cut-off threshold with the water sample and the baseline enhancement gain coefficient ; Step 3.

3. Obtain the modified clipping threshold from the modulation function which is monotonically related to the turbidity factor and an enhanced gain coefficient ; Step 3.4, based on the revised clipping threshold and an enhanced gain coefficient reallocate the pixel histograms of the grid sub-blocks; Step 4, performing high-frequency component extraction and salient region detection on the first intermediate image, and performing gain amplification on the salient edges and textures based on the turbidity factor to obtain a second intermediate image; Step 5, selecting the original underwater image and the second intermediate image for multi-scale decomposition, and performing corresponding fusion after assigning fusion weights according to the turbidity factor and image features at each scale; the specific implementation steps are as follows: Step 5.1, performing multi-scale decomposition on the original underwater image and the second intermediate image respectively to obtain sub-band representations from low frequency to high frequency; Step 5.2, setting a fusion weight factor based on the turbidity factor; determining the fusion weight factor depending on the turbidity factor , using a linear modulation expressed as wherein , ;​ Step 5.3, fusing the corresponding sub-bands of the original underwater image and the second intermediate image at each scale according to the fusion weight factor; Step 6, reconstructing and outputting the fused result to obtain a final enhanced underwater image.

2. The method of claim 1, wherein the method further comprises: In the step 2, the calculation of the turbidity factor further comprises the following steps: Step 2.1, performing scaling on the obtained original underwater image; Step 2.2, simultaneously extracting the image average gradient amplitude, the standard deviation of the image brightness and the average value of the image pixel saturation in the gray domain and the HSV color space; Step 2.3, calculating the turbidity score according to the image average gradient amplitude, the standard deviation of the image brightness and the average value of the image pixel saturation; Step 2.4, normalizing the turbidity score to obtain a turbidity factor corresponding to the actual physical turbidity.

3. The method of claim 1, wherein the method further comprises: The step 4 further comprises the following steps: Step 4.1, performing high-frequency detail extraction on the first intermediate image; Step 4.2, obtaining the edge map of the first intermediate image by using Laplace or Sobel operator; Step 4.3, extracting the fine texture structure of the first intermediate image by high-pass filtering; Step 4.4, introducing visual saliency detection or edge amplitude to screen out weak textures and edges from the edge map and fine texture structure; Step 4.5, applying detail gain to the screened edge map and fine texture structure to obtain the second intermediate image.

4. The method of claim 1, wherein: In step 3.3, the turbidity factor is brought into the increasing function of the monotonic correlation and the decreasing function , based on and corrects the clipping threshold and enhances the gain coefficient ; When the turbidity factor the clipping threshold and the enhancement gain coefficient is close to the reference value; When the turbidity factor is between 0.3 - 0.7, the crop threshold with the enhanced gain coefficient is linearly scaled and amplified; When the turbidity factor the clipping threshold 30% of the reference value, the boost gain factor is boosted up to approximately 1.6 times.

5. The method of claim 3, wherein: In step 4.5, depending on the turbidity factor Controlling the detail gain coefficient , the detail gain coefficient is limited in the interval [0.4, 1.2] and is expressed as wherein , ; At the turbidity factor Near 0, the detail gain coefficient Near 1.2, the texture can be fully highlighted; At turbidity factors The detail gain factor approaches 1 Falls to near 0.4 to suppress false edges due to suspended particles.

6. A system for implementing the method of enhancing the visibility of turbid underwater scenes according to claim 1, characterized in that, It comprises: A turbidity analysis module for obtaining a turbidity factor from an original underwater image by using a turbidimeter or image visual feature estimation processing; A turbidity self-adaptive contrast enhancement module for performing turbidity self-adaptive contrast limited histogram equalization processing on the original underwater image according to the turbidity factor to obtain a first intermediate image; A salient detail enhancement module for performing high-frequency component extraction and salient region detection on the first intermediate image, and performing gain amplification on the salient edges and textures based on the turbidity factor to obtain a second intermediate image; A multi-scale fusion module is configured to perform multi-scale decomposition on the original underwater image and the second intermediate image, and to perform corresponding fusion and reconstruction output after assigning fusion weights according to the turbidity factor and image features at each scale.

Images