A smart optical sensing system and method for imaging underwater turbid environments

Abstract

This application provides an intelligent optical sensing system and method for imaging in turbid underwater environments, belonging to the field of underwater optical imaging. It solves the problem of low efficiency in identifying underwater objects due to the difficulty of obtaining clear images in turbid water with existing equipment. The system includes an optical imaging module, an image acquisition module, and a computing device mounted on a diving device. The computing device runs computer programs for an image preprocessing module, a depth sensing module, and a neural network denoising and recognition module. The optical imaging module generates a structured light field carrying an orbital angular momentum mode based on a spiral phase plate. The image acquisition module acquires the reflection information of the target object after being illuminated by vortex light. The image preprocessing module enhances the acquired image. The depth sensing module obtains the current water pressure using a pressure sensor, and the computing device decomposes the current water depth. The neural network denoising and recognition module performs real-time image recognition and denoising. This application is applied to underwater imaging in turbid water.

Description

Technical Field

[0001] This application relates to the field of underwater optical imaging technology, and in particular to an intelligent optical sensing system and method for imaging through underwater turbid environments. Background Technology

[0002] In underwater search and rescue missions, efficient underwater optical imaging, target identification capabilities, and location information transmission are key technologies for improving search and rescue efficiency and success rates, enabling rapid location and rescue operations. However, in real underwater environments, conventional camera equipment struggles to obtain clear and reliable image information in complex water conditions such as high turbidity, low light, and deep sea, making it difficult for rescuers to accurately identify distressed targets, potential obstacles, or risky areas. This imaging degradation is mainly caused by scattering effects and environmental absorption, and is particularly severe in strongly scattering media. The inability to quickly and accurately obtain environmental information about the distressed area will delay rescue opportunities and may even lead to misjudgments of the environment by rescuers, threatening their own lives.

[0003] Currently, underwater camera-based imaging remains the mainstream method for underwater detection, offering advantages such as high resolution, rapid acquisition, and intuitive visualization. Methods such as multispectral imaging, polarization imaging, binocular vision, and time-gated imaging demonstrate good noise resistance and imaging performance in specific scenarios. These imaging methods can reduce interference from scattering to some extent and improve image clarity. However, existing technologies still have several limitations. First, current imaging systems rely heavily on high-precision components, complex multi-device coordination, and high frame rate acquisition, resulting in high system costs and low integration, making direct deployment in actual search and rescue equipment difficult. Second, in pursuit of high precision and high resolution, many imaging solutions sacrifice real-time imaging performance, failing to meet the urgent need for imaging speed in rescue scenarios. Third, in the underwater environment, the lack of available spatial coordinate information makes it difficult to accurately locate the spatial position of a person in distress, even if successful imaging and identification are achieved, affecting the timeliness and accuracy of rescue deployment. Therefore, there is an urgent need to develop an intelligent optical sensing system that can image through underwater turbid environments, integrating functions such as real-time imaging through underwater turbid environments, target recognition, and real-time location information, and possessing good environmental adaptability and system integrability, so as to meet the technical requirements of "fast, accurate, and safe" for emergency search and rescue missions in turbid waters. Summary of the Invention

[0004] To address the problem of low efficiency in underwater object identification due to the difficulty of obtaining clear images in turbid water with existing equipment, this application proposes an intelligent optical sensing system and method for imaging through turbid underwater environments. Based on vortex illumination and neural network denoising and recognition technology, it utilizes the good transmission properties of vortex light in turbid water, the efficient feature extraction and robust recognition capabilities of neural networks for image data, and the inverse solution algorithm of water pressure-depth to achieve real-time imaging and recognition of target objects and intelligent sensing of their location and depth in complex aquatic environments.

[0005] The technical solution adopted in this application is as follows: an intelligent optical sensing system for imaging through underwater turbid environments, including a computing device and an optical imaging module, an image acquisition module, and a pressure sensor mounted on a diving device. The computing device runs a computer program with an image preprocessing module, a depth sensing module, and a neural network denoising and recognition module. The optical imaging module generates a structured light field carrying an orbital angular momentum mode based on a spiral phase plate. The image acquisition module acquires the reflection information of the target object after being illuminated by vortex light through a high-sensitivity imaging device. The image preprocessing module performs illumination and reflection information separation processing based on the Retinex model, frequency domain Gaussian filtering illumination estimation, and adaptive histogram equalization enhancement on the acquired image. The depth sensing module obtains the current water pressure using the pressure sensor, and the computing device solves for the current water depth. The neural network denoising and recognition module performs real-time recognition and denoising on the image.

[0006] Furthermore, the optical imaging module includes a laser, a beam expander, and a spiral phase plate. The laser emitted by the laser is expanded by the beam expander and then irradiates the spiral phase plate. The beam is modulated to generate vortex light carrying different orbital angular momentum modes, and the vortex beam irradiates the underwater target.

[0007] Furthermore, the image acquisition module includes a lens and a CCD camera. The vortex beam illuminates the underwater target, and the reflected light is collected by the lens and imaged onto the CCD camera. The CCD camera then sends the acquired image to a computing device.

[0008] Furthermore, the neural network denoising and recognition module adopts the data-driven EnhanceNet neural network. The EnhanceNet neural network is an encoder-decoder structure. In the encoding stage, the input image undergoes multiple compression operations to extract image features. Each compression operation passes through two convolutional modules and is downsampled by a max pooling. After the compression operation is completed, the features are flattened into one dimension and input into the attention module. The attention module contains multiple consecutive Transformer layers. Each Transformer layer contains a multi-head self-attention layer and a multi-layer perceptron layer. Both the multi-head self-attention layer and the multi-layer perceptron layer are first processed by layer normalization to stabilize the training process and residual connections are used.

[0009] The decoding part of the network contains two decoders: a denoising decoder and a classification decoder, which perform denoising and classification operations on the input image.

[0010] The denoising decoder gradually restores the spatial resolution of the image through multi-layer upsampling and skip connections. Each decoding stage includes feature concatenation, convolution, non-linear activation and batch normalization. The final output is an image of the same size as the input, achieving noise suppression and detail restoration.

[0011] The classification decoder concatenates the feature information from each layer during the flattening and upsampling process and inputs it into the fully connected layer to output image-level multi-classification. Finally, the network outputs a denoised image and image classification information.

[0012] Furthermore, the diving equipment includes two transparent windows: a light outlet and a light receiver, a pair of horizontal propulsion propellers, a pair of vertical propulsion propellers, and a pressure sensor. The light outlet and light receiver are located at the front of the diving equipment, the pair of horizontal propulsion propellers are symmetrically arranged on both sides of the diving equipment, the pair of vertical propulsion propellers are symmetrically arranged in the middle of the upper and lower surfaces of the diving equipment, and the pressure sensor is located on the upper surface of the diving equipment.

[0013] A smart optical sensing method for imaging underwater turbid environments, based on a smart optical sensing system for imaging underwater turbid environments, includes the following steps:

[0014] Step 1: Illuminate the underwater target using the vortex beam in the optical imaging module;

[0015] Step 2: Acquire the original image of the target object and preprocess it using the image preprocessing module;

[0016] Step 3: Enhance and identify target object images using a data-driven EnhanceNet neural network;

[0017] Step 4: Real-time detection and calculation of water depth.

[0018] Furthermore, step 2 specifically involves:

[0019] Step 2.1: The light reflected from the target object is focused by the lens and imaged on the detection surface of the CCD camera;

[0020] Step 2.2: The original image is acquired by the CCD camera and transmitted to the computing device;

[0021] Step 2.3: Preprocess the raw image received by the computing device. Estimate the collected raw image as the product of the illumination component and the reflection component. Use Gaussian filtering to estimate the illumination information. Then, obtain the illumination information through frequency domain convolution and inverse Fourier transform operations to inversely solve the reflection information. Normalize the reflection information and then process the normalized reflection information through an adaptive histogram equalization algorithm to enhance contrast and improve local details.

[0022] A computer device includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method.

[0023] A computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the method.

[0024] A computer program product includes a computer program / instructions that, when executed by a processor, implement the steps of the method.

[0025] The advantages of this application over the prior art are as follows:

[0026] 1. This application utilizes a vortex beam to illuminate underwater targets, effectively enhancing the anti-interference capability during underwater imaging and increasing the underwater imaging distance. It can effectively detect targets 4m away from the system in turbid underwater environments.

[0027] 2. This application combines neural networks with denoising algorithms, which can adaptively cope with changes in turbidity environment with a large dynamic range, effectively improving the real-time underwater imaging quality by more than 10dB, providing rescuers with clearer and more reliable visual and distance information, and improving the efficiency and success rate of rescue.

[0028] 3. This application combines structured light illumination with data-driven intelligent recognition methods, which significantly improves the recognition accuracy of dynamic targets in turbid environments and is applicable to a variety of complex application scenarios such as underwater search and rescue, water security, and water environment monitoring. Attached Figure Description

[0029] The following description, in conjunction with the accompanying drawings, further illustrates this application:

[0030] Figure 1 A schematic diagram of the overall system structure provided for embodiments of this application;

[0031] Figure 2 This is a schematic diagram of the optical imaging structure provided in an embodiment of this application;

[0032] Figure 3 This is a schematic diagram of the structure of the image denoising and recognition network provided in the embodiments of this application;

[0033] Figure 4 Imaging experimental results provided for verification of the validity of this application in the embodiments of this application;

[0034] Figure 5 The imaging experimental results quality graph provided for the embodiments of this application to verify the validity of this application;

[0035] Figure 6 Dynamic imaging experimental results provided for verification of the validity of this application in the embodiments of this application;

[0036] In the diagram: 101 is the light output port, 102 is the light receiving port, 103 is the horizontal propulsion propeller, 104 is the vertical propulsion propeller, 105 is the pressure sensor, 201 is the laser, 202 is the beam expander, 203 is the spiral phase plate, 204 is the target object, 205 is the lens, 206 is the CCD camera, and 207 is the computing device. Detailed Implementation

[0037] like Figures 1 to 6 As shown, this application provides an intelligent optical sensing system for imaging through underwater turbid environments. The system includes an optical imaging module, an image acquisition module, a pressure sensor, and a computing device 207. The optical imaging module, image acquisition module, and pressure sensor are mounted on a diving device. The computing device 207 runs computer programs for an image preprocessing module, a depth sensing module, and a neural network denoising and recognition module. The optical imaging module, based on a spiral phase plate, generates a structured light field carrying an orbital angular momentum mode, enhancing its anti-scattering capability in underwater turbid environments. The image acquisition module acquires the reflection information of the target object after vortex light illumination through a high-sensitivity imaging device and sends it to the computing device 207. The image preprocessing module performs Retinex model-based illumination and reflection information separation processing, frequency domain Gaussian filtering illumination estimation, and adaptive histogram equalization enhancement on the acquired image. The depth sensing module obtains the current water pressure using the pressure sensor 105, and the computing device 207 decomposes the current water depth. The neural network denoising and recognition module adopts an encoder-decoder structure, combining feature learning and pattern matching to perform real-time image recognition and denoising.

[0038] This diving device integrates an optical imaging module, combining the vortex beam illumination component with a neural network. This enables effective capture of target scene images in highly turbid water while utilizing a deep neural network for image denoising and target recognition, significantly improving the system's recognition distance in turbid water environments. Furthermore, by adding an external pressure sensor and employing an inverse kinematics algorithm, the system can achieve depth localization of underwater targets. This application addresses the technical problems of poor imaging quality, low target recognition accuracy, and inability to accurately locate targets in turbid water environments.

[0039] like Figure 1As shown in the figure, this embodiment illustrates the structure of an implementable diving device. This diving device involves two transparent windows: a light-emitting port 101 and a light-receiving port 102, a pair of horizontal propellers 103, a pair of vertical propellers 104, and a pressure sensor 105. The light-emitting port 101 and the light-receiving port 102 are located at the front of the diving device. The pair of horizontal propellers 103 are symmetrically arranged on both sides of the diving device, and the pair of vertical propellers 104 are symmetrically arranged in the middle of the upper and lower surfaces of the diving device. The pressure sensor 105 is located on the upper surface of the diving device. Furthermore, a communication module, including wireless and wired communication modules, can be integrated into the diving device to transmit the acquired images to a computing device 207, which acts as a host computer.

[0040] like Figure 2 As shown, the optical imaging module includes a laser 201. The laser emitted by the laser 201 is expanded by a beam expander 202 and then irradiates a spiral phase plate 203, where it is modulated into a vortex beam. The vortex beam irradiates an underwater target 204, and the reflected light is collected by a lens 205 and imaged onto a CCD camera 206. The image is transmitted to a computing device 207, preprocessed, and then transmitted to a neural network to output a denoised image and target recognition information. A pressure sensor 105 collects pressure information in real time and transmits it to the computing device 207 to display the water depth where the diving equipment is located in real time.

[0041] like Figure 3 As shown, the neural network denoising and recognition module adopts the data-driven EnhanceNet neural network. The EnhanceNet neural network is an encoder-decoder structure. In the encoding stage, the input image undergoes multiple compression operations to extract image features. Each compression operation passes through two convolutional modules and is downsampled by a 2×2 max pooling layer. The convolutional module contains a 3×3 convolution and a ReLU activation function. After compression, the features are flattened into one dimension and input into the attention module, which contains eight consecutive Transformer layers. Each Transformer layer contains a multi-head self-attention layer (MSA) and a multilayer perceptron layer (MLP). Both the MSA and MLP undergo layer normalization to stabilize the training process and use residual connections. The addition of the attention mechanism enables global, multi-angle, and deep feature information extraction, improving the robustness of the neural network.

[0042] The denoising decoder gradually restores the spatial resolution of the image through multiple upsampling layers and skip connections. Each decoding stage includes feature concatenation, convolution, nonlinear activation, and batch normalization, ultimately outputting an image of the same size as the input. This achieves noise suppression and detail restoration. The network uses mean squared error loss as its loss function.

[0043]

[0044] In the formula, y i Let i be the true value of the i-th pixel. Let be the predicted value of the i-th pixel, and N be the total number of pixels in the image.

[0045] The classification decoder concatenates the feature information from each layer during the flattening and upsampling process and inputs it into the fully connected layer. The fully connected layer contains three hidden layers and one output layer. The output of the output layer is a vector of length 10, representing the class probability distribution. Finally, the network outputs a denoised image and classification information.

[0046] Based on the above system, this application also proposes an intelligent optical sensing method for imaging underwater turbid environments, comprising the following steps:

[0047] Step 1: Illuminate the underwater target 204 with the vortex beam in the optical imaging module, specifically as follows:

[0048] Step 1.1: Place the fish model as target object 204 in turbid water;

[0049] Step 1.2: The 532nm wavelength laser emitted by the laser 201 is expanded by the beam expander 202 and then modulated by the spiral phase plate 203 to generate vortex light carrying different orbital angular momentum modes, which then illuminates the target object 204.

[0050] Step 2: Acquire the original image of the target object 204 and preprocess it using the image preprocessing module;

[0051] Step 2.1: The light reflected by the target object 204 is converged by the lens 205 and imaged on the detection surface of the CCD camera 206;

[0052] Step 2.2: The original image is acquired by the CCD camera 206 and transmitted to the computing device 207;

[0053] Step 2.3: Preprocess the original image received by the computing device 207, estimate the collected original image as the product of the illumination component and the reflection component, estimate the illumination information using Gaussian filtering, and then obtain the illumination information through operations such as frequency domain convolution and inverse Fourier transform for inverse decomposition of reflection information; normalize the reflection information, and then process the normalized reflection information through an adaptive histogram equalization algorithm to enhance contrast and improve local details.

[0054] The specific calculation process of preprocessing is as follows: The Retinex algorithm is used to separate the illumination information and recover the reflection image. The image information modeled in the Retinex algorithm is as follows:

[0055] I(x,y)=L(x,y)·R(x,y);

[0056] In the formula, I(x,y) represents the original image, L(x,y) represents the illumination component, and R(x,y) represents the reflection component;

[0057] Step 2.4: Estimate lighting information using Gaussian filtering:

[0058]

[0059] In the formula, G σ (x,y) represents the lighting information, and σ is the standard deviation, which determines the smoothness of the lighting information.

[0060] Perform frequency domain convolution and inverse Fourier transform to obtain illumination information:

[0061]

[0062] In the formula, For the spectral information of lighting, Indicates Fourier transform;

[0063] Step 2.5: Estimate reflection information from the image modeling in Step 2.3:

[0064] logR(x,y)=logI(x,y)-logL(x,y);

[0065] After restoring the brightness range, the reflection information is normalized:

[0066] R′(x,y)=exp(logR(x,y));

[0067]

[0068] In the formula, R′(x,y) represents the reflection information after restoring the brightness range, and R″(x,y) represents the normalized reflection information;

[0069] Step 2.6: Use the adaptive histogram equalization algorithm from the MATLAB library to process the normalized reflection information, further enhancing contrast and improving local details:

[0070] I0(x,y)=adapthisteq(R″(x,y));

[0071] In the formula, I0(x,y) is the enhanced image in the preprocessing process, and adapthisteq is the adaptive histogram equalization algorithm.

[0072] Step 3: Enhance and recognize target images based on data-driven EnhanceNet neural network. Specifically, the enhanced image obtained in step 2.6 is processed using EnhanceNet neural network to obtain a denoised image and classify the target in the image.

[0073] Step 4: Real-time water depth detection, specifically:

[0074] Step 4.1: The pressure sensor 105 collects the pressure in real time, returns the voltage value, and transmits it to the computing device 207;

[0075] Step 4.2: The computing device 207 determines the underwater pressure through the sensor calibration relationship, and then determines the current sensor depth using the pressure formula:

[0076] P = k·V + b;

[0077]

[0078] In the formula, P is the underwater pressure, V is the voltage value returned by the sensor, k is the sensor sensitivity coefficient, b is the zero drift, k and b are calibrated by water pressure before the experiment, ρ is the water density, and P atm The current atmospheric pressure is calibrated before the experiment; the program converts the returned voltage value into water depth h in real time and outputs image annotations in real time.

[0079] The validity of this application is verified through experiments below:

[0080] To verify the imaging performance of this system under different turbidities and target distances (i.e., the distance between the target and the system), a corresponding experimental environment was constructed. In the experiment, the system position was kept fixed (1 meter above the water surface at a depth of 1 meter), and tests were conducted in a 4-meter-long water tank by adjusting the water turbidity (range 10 to 80 NTU, target distance 1 meter) and the target distance (range 0.5 to 4 meters, water turbidity 15 NTU). To ensure consistency under different experimental conditions, the focal length was adjusted during the experiment to maintain a consistent target imaging size.

[0081] Through analysis Figure 4-6 We can conclude that:

[0082] (1) As Figure 4As shown, an underwater environment with turbidity increasing from 10 NTU to 80 NTU was constructed with a target distance of 1 meter. It can be seen that as the turbidity increases, the original image changes from clear to blurry, and the target object becomes indistinguishable at 80 NTU. However, although the quality of the optimized output image also gradually decreases, the target object can still be clearly identified by the human eye in an underwater environment with a turbidity of 80 NTU. Furthermore, with a water turbidity of 15 NTU, the imaging results of underwater targets were measured sequentially as the target distance increased from 0.5 meters to 4 meters. It can be seen that as the distance between the target object and the system increases, the original image gradually changes from clear to blurry, and the target object becomes indistinguishable at a distance of 3 meters. Conversely, although the quality of the optimized output image also gradually decreases, the target object can still be clearly identified by the human eye in an underwater environment with a distance of 4 meters.

[0083] The results show that this system can not only cope with strong scattering and absorption caused by target distance, but also with scattering noise caused by turbidity, effectively suppressing noise in the original images obtained directly by CCD camera.

[0084] 2) such as Figure 5 As shown, the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) parameters of all images in underwater turbid environments (0-80 NTU) and at target distances (0.25-4 meters) were calculated.

[0085] At a fixed target distance, as turbidity increases, the PSNR value of the original image drops sharply from 35dB to 27dB, then gradually decreases to 20dB within the range of 20-80 NTU. Similarly, the SSIM value of the original image drops sharply from 0.9 to 0.28. Conversely, for the output image optimized by the system, although both PSNR and SSIM show a decreasing trend, the decrease is slow. Its PSNR value gradually decreases from 35dB to 32dB, and the SSIM value decreases by 0.05.

[0086] Furthermore, at a constant underwater turbidity, as the target distance increased, the PSNR value of the original image dropped sharply from 33dB to 25dB, then gradually decreased to 16dB, and its SSIM value dropped sharply from 0.85 to 0.33, then gradually decreased to 0.18. Conversely, for the output image after system optimization, although the target distance had some impact on image quality, the PSNR decrease was not significant, dropping from 33dB to 27dB, while the SSIM value only decreased by 0.08.

[0087] The results show that in underwater environments with high turbidity and a large distance between the target and the system, the image quality improved by more than 10 dB after optimization by this system.

[0088] 3) such as Figure 6As shown, an image set was formed by periodically extracting frames from the dynamic video to analyze and describe the imaging performance of this system during dynamic processes (target distance 1.25 meters, turbidity 25 NTU). It can be observed that, under underwater turbidity conditions, the system output image of the dynamic target is of stable quality, with no obvious artifacts or fluctuations. The experimental results verify the reliability and stability of the system in practical application scenarios.

[0089] 4) Five hundred scenarios without targets and one thousand scenarios containing targets were collected to verify the system's effective recognition capability. Experimental results showed that only 17 scenarios without targets were falsely reported, and 15 scenarios containing targets were missed. Calculations showed that the false alarm rate was as low as 1.7%, and the false alarm rate was as low as 3.0%. Furthermore, the density of the water was measured to be 1028 kg / m³. 3 The returned voltage value is 1.01V, and the calculated water depth of the system is 1.05m.

[0090] In summary, this application proposes for the first time a high-scattering medium imaging and localization scheme based on structured light illumination and network denoising, and proposes a real-time underwater target identification scheme with low false alarm rate, providing strong technical support for emergency command and underwater rescue.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. An intelligent optical sensing system for imaging through underwater turbid environments, characterized in that: The system includes a computing device and an optical imaging module, an image acquisition module, and a pressure sensor mounted on the diving equipment. The computing device runs a computer program containing an image preprocessing module, a depth sensing module, and a neural network denoising and recognition module. The optical imaging module generates a structured light field carrying an orbital angular momentum mode based on a spiral phase plate. The image acquisition module acquires the reflection image of the target object after being illuminated by vortex light through a high-sensitivity imaging device and transmits it to the computing device. The image preprocessing module performs illumination and reflection information separation processing based on the Retinex model, frequency domain Gaussian filtering illumination estimation, and adaptive histogram equalization enhancement on the acquired image. The depth sensing module obtains the current water pressure using the pressure sensor, and the computing device solves for the current water depth. The neural network denoising and recognition module performs real-time image recognition and denoising. The neural network denoising and recognition module adopts the data-driven EnhanceNet neural network. The EnhanceNet neural network is an encoder-decoder structure. In the encoding stage, the input image undergoes multiple compression operations to extract image features. Each compression operation passes through two convolutional modules and is downsampled by a max pooling. After the compression operation is completed, the features are flattened into one dimension and input into the attention module. The attention module contains multiple consecutive Transformer layers. Each Transformer layer contains a multi-head self-attention layer and a multi-layer perceptron layer. Both the multi-head self-attention layer and the multi-layer perceptron layer are first processed by layer normalization to stabilize the training process and residual connections are used. The EnhanceNet neural network decoding part contains two decoders: a denoising decoder and a classification decoder, which perform denoising and classification operations on the input image. The denoising decoder gradually restores the spatial resolution of the image through multi-layer upsampling and skip connections. Each decoding stage includes feature concatenation, convolution, non-linear activation and batch normalization. The final output is an image of the same size as the input, achieving noise suppression and detail restoration. The classification decoder concatenates the feature information from each layer during the flattening and upsampling process and inputs it into the fully connected layer to output image-level multi-classification. Finally, the network outputs a denoised image and image classification information.

2. The intelligent optical sensing system for imaging through underwater turbid environments according to claim 1, characterized in that: The optical imaging module includes a laser, a beam expander, and a spiral phase plate. The laser emitted by the laser is expanded by the beam expander and then shines on the spiral phase plate. The beam is modulated to generate vortex light carrying different orbital angular momentum modes, and the vortex beam shines on the underwater target.

3. The intelligent optical sensing system for imaging through underwater turbid environments according to claim 2, characterized in that: The image acquisition module includes a lens and a CCD camera. A vortex beam illuminates the underwater target, and the reflected light is collected by the lens and imaged onto the CCD camera. The CCD camera then sends the acquired image to a computing device.

4. The intelligent optical sensing system for imaging through underwater turbid environments according to any one of claims 1-3, characterized in that: The diving equipment includes two transparent windows: a light outlet and a light receiver, a pair of horizontal propellers, a pair of vertical propellers, and a pressure sensor. The light outlet and light receiver are located at the front of the diving equipment. The pair of horizontal propellers are symmetrically arranged on both sides of the diving equipment. The pair of vertical propellers are symmetrically arranged in the middle of the upper and lower surfaces of the diving equipment. The pressure sensor is located on the upper surface of the diving equipment.

5. An intelligent optical sensing method for imaging through underwater turbid environments, based on the intelligent optical sensing system for imaging through underwater turbid environments as described in any one of claims 1-4, characterized in that: Includes the following steps: Step 1: Illuminate the underwater target using the vortex beam in the optical imaging module; Step 2: Acquire the original image of the target object and preprocess it using the image preprocessing module; Step 3: Enhance and identify target object images using a data-driven EnhanceNet neural network; Step 4: Real-time detection and calculation of water depth.

6. The intelligent optical sensing method for imaging through underwater turbid environments according to claim 5, characterized in that: Step 2 is as follows: Step 2.1: The light reflected from the target object is focused by the lens and imaged on the detection surface of the CCD camera; Step 2.2: The original image is acquired by the CCD camera and transmitted to the computing device; Step 2.3: Preprocess the raw image received by the computing device. Estimate the collected raw image as the product of the illumination component and the reflection component. Use Gaussian filtering to estimate the illumination information. Then, obtain the illumination information through frequency domain convolution and inverse Fourier transform operations to inversely solve the reflection information. Normalize the reflection information and then process the normalized reflection information through an adaptive histogram equalization algorithm to enhance contrast and improve local details.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method of claim 5 or 6.

8. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that: When the computer program / instructions are executed by the processor, they implement the steps of the method described in claim 5 or 6.

9. A computer program product, comprising a computer program / instructions, characterized in that: When the computer program / instructions are executed by the processor, they implement the steps of the method described in claim 5 or 6.

Images