Image segmentation method and system for stone removal

Abstract

The present invention relates to the field of image analysis technology and discloses an image segmentation method and system for stone removal. The method comprises: performing real-time image acquisition of a target area based on a pre-deployed stone image acquisition device to obtain an initial stone image; preprocessing the initial stone image to obtain a preprocessed initial stone image; inputting the preprocessed initial stone image into a deep learning model based on a convolutional neural network to obtain a preliminary segmentation result; optimizing the preliminary segmentation result using a Sobel operator combined with a closed operation to obtain an optimized segmentation result; and applying a connected domain analysis method to the optimized segmentation result to obtain a complete stone region image. The present invention can solve the problem in current technologies where seed point selection is affected by noise, resulting in inaccurate segmentation results.

Description

Technical Field

[0001] The present invention relates to the field of image analysis technology, and in particular to an image segmentation method and system for stone removal. Background Art

[0002] With the continuous advancement of modern medical imaging technology, the diagnosis and treatment of stones has gradually shifted from traditional manual diagnosis to digital and automated computer-assisted diagnosis (CAD) systems. Image segmentation technology plays a crucial role in stone removal, particularly in extracting the stone area and assisting doctors in making accurate diagnoses and treatment decisions. Stone image segmentation often faces multiple challenges, including blurred boundaries between the stone and surrounding tissue, noise interference, and diverse stone shapes.

[0003] Traditional image segmentation methods, such as threshold segmentation, edge detection, and region growing, while effective in certain simple scenarios, are less than ideal due to background noise in stone images, the complexity of stone morphology, and image variations under different imaging conditions. With the successful application of deep learning, particularly convolutional neural networks (CNNs), in computer vision, deep learning methods have gradually become the mainstream approach for medical image segmentation. However, existing deep learning-based image segmentation methods still suffer from issues such as inaccurate stone region extraction, poor recognition of complex stone morphologies, and high computational resource consumption. In particular, edge information extraction and feature fusion processing often rely on complex network structures, resulting in low computational efficiency.

[0004] Therefore, developing an efficient and accurate stone image segmentation method that can retain details and reduce noise interference when extracting stone areas and can cope with stones of different shapes and sizes is an urgent problem to be solved in the current field of medical image processing. Summary of the Invention

[0005] In view of this, the present invention proposes an image segmentation method and system for stone removal, aiming to solve the problem in current technology that seed point selection is affected by noise, resulting in inaccurate segmentation results.

[0006] The present invention proposes an image segmentation method for stone removal, comprising:

[0007] Performing real-time image acquisition of the target area based on a pre-deployed stone image acquisition device to obtain an initial stone image, and preprocessing the initial stone image to obtain a preprocessed initial stone image;

[0008] The preprocessed initial stone image is input into a deep learning model based on a convolutional neural network to obtain preliminary segmentation results;

[0009] The preliminary segmentation results are optimized using the Sobel operator combined with the closing operation to obtain the optimized segmentation results;

[0010] The connected domain analysis method is used on the optimized segmentation results to obtain a complete stone area image.

[0011] Furthermore, the preprocessing of the initial stone image to obtain the preprocessed initial stone image specifically comprises: performing noise elimination on the initial stone image using a multi-scale filtering method to obtain a denoised stone image, and performing contrast enhancement and histogram equalization processing on the denoised stone image to obtain a preprocessed initial stone image.

[0012] Furthermore, the preprocessed initial stone image is input into a deep learning model based on a convolutional neural network to obtain a preliminary segmentation result. The specific content is: the preprocessed initial stone image is input into a deep learning model based on a convolutional neural network, the local edge information of the preprocessed initial stone image is extracted through the shallow features in the encoder, and the global context information of the preprocessed initial stone image is extracted through the deep features. Finally, the shallow features and the deep features are fused through jump connections and decoded by the decoder to obtain a confidence map. By setting a threshold for the confidence map, the area in the confidence map that is greater than the set threshold is automatically extracted as a seed point, and the preliminary segmentation result is obtained by combining the region growing method.

[0013] Furthermore, the convolutional neural network-based deep learning model adopts an improved U-Net architecture, that is, the residual module, parallel convolution module and attention module are combined on the original U-Net architecture. The improved U-Net architecture includes an encoder and a decoder. The encoder includes multiple convolution modules and pooling layers. Each convolution module in the multiple convolution modules is added with a residual module, and finally a parallel convolution module is added. The decoder includes a deconvolution module, a feature fusion module, a convolution module and a parallel convolution module. The feature fusion module includes an attention module.

[0014] Furthermore, the local edge information of the preprocessed initial stone image is extracted through shallow features in the encoder, specifically as follows: the preprocessed initial stone image is input into the encoder, normalized and then subjected to a convolution operation with a 3×3 convolution kernel, nonlinear expression capability is introduced through an activation function, and then convolution operations with multiple 3×3 convolution kernels in the convolution module are used to extract basic edge information, texture, and low-level features in the preprocessed initial stone image, normalize the basic edge information, texture, and low-level features in the preprocessed initial stone image, perform maximum pooling processing on the basic edge information, texture, and low-level features in the preprocessed initial stone image, and then form local edge information through splicing through a parallel convolution module;

[0015] The basic edge information includes boundary lines and contour lines in the pre-processed initial stone image;

[0016] The texture includes the roughness of the stone surface and the degree of concavity and convexity of the stone surface;

[0017] The low-level features include grayscale contrast and color contrast between the stone and the pre-processed initial stone image background.

[0018] Furthermore, the global context information of the preprocessed initial stone image is extracted through deep features. Specifically, the preprocessed initial stone image is input into the encoder, normalized and then subjected to a convolution operation with a 3×3 convolution kernel, nonlinear expression capability is introduced through an activation function, and then multiple 3×3 convolution operations in multiple convolution modules are performed to extract local features in the preprocessed initial stone image. After each convolution operation, maximum pooling processing is performed through the pooling layer to aggregate the local features into global context information. The global context information obtained after each convolution operation is integrated through a parallel convolution module to obtain the global context information of the preprocessed initial stone image.

[0019] Furthermore, the shallow features and deep features are finally fused through jump connections and decoded by the decoder to obtain a confidence map. The specific content is: the shallow features and deep features are input into the decoder, first passed through the deconvolution module, and the inverse convolution operation is performed, and then a jump connection is performed to fuse and splice the shallow features and deep features that have passed the deconvolution module to obtain a fused feature map, and then the fused feature map is convolved, and the attention module is combined to improve the accuracy of the fused feature map. Finally, the fused feature map with improved accuracy is mapped to a single-channel confidence map through a 1×1 convolution layer.

[0020] Furthermore, the preliminary segmentation result is optimized using the Sobel operator combined with a closing operation to obtain an optimized segmentation result. The specific content is: the Sobel operator is used to perform edge detection on the preliminary segmentation result, and the edge information of the preliminary segmentation result is extracted by calculating the gradient of the image to obtain the stone area. The stone area is then dilated to expand the stone area, and finally an erosion operation is performed to restore the original shape of the stone area and remove the noise introduced in the dilation operation, thereby finally forming an optimized segmentation result.

[0021] Furthermore, the connected domain analysis method is used on the optimized segmentation results to obtain a complete stone area image. The specific content is: the optimized segmentation results are divided into regions based on pixel values, and the same pixel values ​​are classified into the same stone area, thereby converting the optimized segmentation results into a complete stone area image.

[0022] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By adopting an improved U-Net architecture combined with a residual module, a parallel convolution module and an attention module, it is possible to effectively extract shallow and deep features in the image, especially in capturing edge information and fusing multi-scale features. The use of jump connections further enhances the fusion of shallow and deep features and improves the fineness of the image segmentation results. (2) The Sobel operator is combined with a closed operation to optimize the preliminary segmentation results, which can further refine the segmentation boundaries and enhance the coherence of the stone area. The combination of dilation and erosion operations effectively removes noise and artifacts, ensuring the accurate extraction of the stone area. (3) The optimized segmentation results are post-processed by the connected domain analysis method to ensure that the complete stone area image can be extracted, avoiding the omission or misidentification of the segmented area, and providing accurate regional positioning for subsequent stone removal and medical diagnosis. (4) Through the improved U-Net architecture, the powerful feature extraction capabilities of deep convolutional neural networks are fully utilized. At the same time, combined with lightweight network optimization, the model has effectively improved its computational efficiency while ensuring high accuracy, making it suitable for actual clinical applications. (5) It can automatically perform image acquisition, preprocessing, segmentation and post-processing, greatly reducing the need for manual intervention, reducing the workload of doctors, and improving diagnostic efficiency. In addition, the system can automatically extract stone areas based on the set thresholds and seed points, further improving the automation level of the system.

[0023] On the other hand, the present invention also provides an image segmentation system for stone removal, the system comprising:

[0024] The first module is configured to perform real-time image acquisition of the target part based on a pre-deployed stone image acquisition device to obtain an initial stone image, and pre-process the initial stone image to obtain a pre-processed initial stone image;

[0025] The second module is configured to input the preprocessed initial stone image into a deep learning model based on a convolutional neural network to obtain a preliminary segmentation result;

[0026] The third module is configured to optimize the preliminary segmentation result using the Sobel operator combined with the closing operation to obtain the optimized segmentation result;

[0027] The fourth module is configured to use the connected domain analysis method on the optimized segmentation results to obtain a complete stone area image.

[0028] It is understandable that the above-mentioned image segmentation system for stone removal has the same beneficial effects and will not be described in detail here. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0030] Figure 1 This is a flow chart of an image segmentation method for stone removal according to an embodiment of the present invention;

[0031] Figure 2 This is a structural block diagram of an image segmentation system for stone removal according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, unless there is a conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0033] See Figure 1 As shown, an embodiment of the present invention provides an image segmentation method for stone removal, comprising:

[0034] S1: Based on the pre-deployed stone image acquisition device, real-time image acquisition is performed on the target area to obtain an initial stone image, and the initial stone image is preprocessed to obtain a preprocessed initial stone image;

[0035] S2: The preprocessed initial stone image is input into the deep learning model based on convolutional neural network to obtain the preliminary segmentation result;

[0036] S3: The preliminary segmentation result is optimized using the Sobel operator combined with the closing operation to obtain the optimized segmentation result;

[0037] S4: The connected domain analysis method is used on the optimized segmentation results to obtain a complete stone area image.

[0038] Furthermore, the initial stone image is preprocessed to obtain the preprocessed initial stone image. Specifically, the initial stone image is subjected to noise elimination using a multi-scale filtering method to obtain a denoised stone image, and the denoised stone image is subjected to contrast enhancement and histogram equalization processing to obtain a preprocessed initial stone image.

[0039] It should be noted that the multi-scale filtering method can effectively remove different types of noise (such as salt and pepper noise, artifacts, etc.) by analyzing images at multiple scales, thereby improving image quality. By increasing the contrast of the image, the outline and details of the stone will be more prominent, thereby improving the segmentation algorithm's ability to recognize the stone area. Especially in low-contrast areas, histogram equalization can effectively enhance the details of the stone area, improve the contrast between the stone and the background, and make the stone boundary more distinct.

[0040] Furthermore, the preprocessed initial stone image is input into a deep learning model based on a convolutional neural network to obtain a preliminary segmentation result. The specific content is: the preprocessed initial stone image is input into a deep learning model based on a convolutional neural network, the local edge information of the preprocessed initial stone image is extracted through the shallow features in the encoder, and the global context information of the preprocessed initial stone image is extracted through the deep features. Finally, the shallow features and the deep features are fused through jump connections and decoded by the decoder to obtain a confidence map. By setting a threshold for the confidence map, the area in the confidence map that is greater than the set threshold is automatically extracted as a seed point, and the preliminary segmentation result is obtained by combining the region growing method.

[0041] Specifically, a confidence map is a graph that reflects the probability that each pixel belongs to a stone region. Each pixel value in the map represents the probability that the pixel belongs to a stone region. Larger values ​​indicate a higher probability of a stone region, while smaller values ​​indicate a background or non-stone region.

[0042] Based on actual application requirements and image quality, an appropriate threshold is set for the confidence map. This threshold is typically an empirical value or an optimal value tuned based on training data. After the threshold is set, regions in the confidence map with pixel values ​​greater than the threshold are selected as "seed points." These seed points represent pixels with a high probability of being stone areas.

[0043] Based on a set threshold, the system automatically extracts all regions with values ​​greater than the threshold from the confidence map as seed points. These seed points serve as the starting points for region growing and have a high probability of containing stones. This allows the system to automatically identify possible stone regions within the entire image, eliminating manual labeling and intervention.

[0044] After extracting the seed point, the region growing algorithm is used to expand the pixels surrounding the seed point. This method is based on the similarity of the local area, comparing adjacent pixels with the seed point. If the values ​​of adjacent pixels are also greater than a set threshold, these pixels are added to the stone region. Through continuous expansion, the region growing method gradually constructs the entire stone region.

[0045] Finally, the region obtained by the region growing algorithm is the preliminary segmentation result. In this result, the stone region has been successfully segmented from the background. The gradual expansion of region growing ensures the coherence of the stone region while avoiding missegmentation caused by image noise or blurred edges.

[0046] It should be noted that by automatically extracting seed points through threshold setting and combining them with the region growing method, the entire segmentation process significantly reduces manual intervention. This automated processing method can significantly improve segmentation efficiency, reduce errors caused by human factors, and enhance segmentation consistency. After setting an appropriate threshold and extracting seed points, the stone area can be accurately located. The threshold setting helps the system filter out low-probability noise areas, ensuring more reliable segmentation results. Seed points are extracted based on high-probability values ​​in the confidence map, avoiding incorrect segmentation.

[0047] Furthermore, an improved U-Net architecture is adopted based on the convolutional neural network deep learning model, that is, the residual module, parallel convolution module and attention module are combined with the original U-Net architecture. The improved U-Net architecture includes an encoder and a decoder. The encoder includes multiple convolution modules and pooling layers. Each convolution module in the multiple convolution modules is added with a residual module, and finally a parallel convolution module is added. The decoder includes a deconvolution module, a feature fusion module, a convolution module and a parallel convolution module. The feature fusion module includes an attention module.

[0048] It should be noted that residual connections avoid the problem of vanishing gradients when the network is deepened, thereby accelerating the convergence process and improving the stability of training. Residual connections allow information to be transferred between different layers, preventing information from being weakened during multi-layer transmission, enhancing the network's representation ability and ability to learn complex features. The residual module optimizes information flow through cross-layer connections, enabling the network to extract more profound features and reducing model degradation during training.

[0049] The parallel convolution module can process convolution kernels of different sizes in parallel at the same level, capturing information at more scales and accurately identifying stones of different sizes and shapes. The combination of multiple convolution kernels at the same level enriches the network's understanding of details and overall structure, thereby improving segmentation. Through parallel convolution, features of different scales can be fully integrated, further improving segmentation accuracy, especially in the processing of stone boundaries and details.

[0050] The attention module can enable the network to pay more attention to the key information of stones (such as edges, contours, etc.) during decoding, reduce the impact of background noise, and thus optimize the image segmentation results. By weighting important areas, the attention mechanism can dynamically adjust the focus of the network, making the network's segmentation of stone areas more accurate, especially in cases of complex morphology or poor image quality. By selectively focusing on important features, the attention module helps the network reduce computational redundancy and improve computational efficiency, while improving the network's generalization ability and avoiding overfitting of irrelevant information.

[0051] Furthermore, the local edge information of the preprocessed initial stone image is extracted through shallow features in the encoder. Specifically, the preprocessed initial stone image is input into the encoder, normalized, and then convolved with a 3×3 convolution kernel. Nonlinear expression capability is introduced through the activation function. The basic edge information, texture, and low-level features of the preprocessed initial stone image are extracted through convolution operations with multiple 3×3 convolution kernels in the convolution module. The basic edge information, texture, and low-level features of the preprocessed initial stone image are normalized and then subjected to maximum pooling in the pooling layer. The local edge information is then formed by splicing through the parallel convolution module.

[0052] Basic edge information includes boundary lines and contour lines in the preprocessed initial stone image;

[0053] Texture includes the roughness of the stone surface and the degree of concavity and convexity of the stone surface;

[0054] The low-level features include the grayscale contrast and color contrast between the stone and the pre-processed initial stone image background.

[0055] Furthermore, the global context information of the preprocessed initial stone image is extracted through deep features. Specifically, the preprocessed initial stone image is input into the encoder, normalized and then subjected to a convolution operation with a 3×3 convolution kernel. Nonlinear expression ability is introduced through the activation function, and then multiple 3×3 convolution operations in multiple convolution modules are used to extract local features in the preprocessed initial stone image. After each convolution operation, the maximum pooling process is performed through the pooling layer to aggregate the local features into the global context information. The global context information obtained after each convolution operation is integrated through the parallel convolution module to obtain the global context information of the preprocessed initial stone image.

[0056] Furthermore, the shallow features and deep features are fused through jump connections and decoded by the decoder to obtain a confidence map. The specific content is: the shallow features and deep features are input into the decoder, first pass through the deconvolution module, perform inverse convolution operation, and then perform jump connections to fuse and splice the shallow features and deep features that have passed the deconvolution module to obtain a fused feature map, and then perform convolution operation on the fused feature map. At the same time, the attention module is used to improve the accuracy of the fused feature map. Finally, the fused feature map with improved accuracy is mapped to a single-channel confidence map through a 1×1 convolution layer.

[0057] Furthermore, the preliminary segmentation results are optimized using the Sobel operator combined with a closing operation to obtain the optimized segmentation results. Specifically, the Sobel operator is used to perform edge detection on the preliminary segmentation results, and the edge information of the preliminary segmentation results is extracted by calculating the gradient of the image to obtain the stone area. Subsequently, a dilation operation is performed on the stone area to expand the stone area. Finally, an erosion operation is performed to restore the original shape of the stone area and remove the noise introduced in the dilation operation, ultimately forming the optimized segmentation result.

[0058] Specifically, the Sobel operator is used to perform edge detection on the initial segmentation results. The Sobel operator calculates the gradient of each pixel in the image to extract edge information from local regions. When processing stone images, the Sobel operator can effectively capture the boundary between the stone and surrounding tissue, thereby clarifying the location and morphological characteristics of the stone area. Through horizontal and vertical convolution operations, the Sobel operator can reveal changes in edge intensity in the image, thereby helping to distinguish stones from the background.

[0059] Dilation is performed on the stone region extracted using the Sobel operator to expand the stone region. Dilation expands the pixels in the surrounding neighborhood to fill small gaps or discontinuities within the stone region, ensuring its continuity. Dilation can compensate for details that may be lost during edge detection, making the stone region more complete and eliminating small areas of missed detection or missing information due to image noise.

[0060] After the dilation operation, the erosion operation is performed. Erosion is the inverse of dilation, restoring the original shape of the stone region by shrinking the boundaries of the dilated region. Erosion removes noise and artifacts that may have been introduced by the dilation operation, particularly irrelevant pixels that were mistakenly extended into the background during dilation. Erosion restores the shape of the stone region to a more realistic state while removing small-scale artifacts caused by noise or errors, improving the accuracy of the segmentation results.

[0061] After edge detection using the Sobel operator, followed by dilation and erosion, the optimized segmentation result is finally obtained. This more accurately extracts the stone region, avoiding artifacts or missed segments that may have existed in the initial segmentation results. The stone region's boundaries are clearer, and its morphology is more consistent with the actual situation.

[0062] Furthermore, the connected domain analysis method is used on the optimized segmentation results to obtain a complete stone area image. The specific content is: the optimized segmentation results are divided into regions based on pixel values, and the same pixel values ​​are classified into the same stone area, thereby converting the optimized segmentation results into a complete stone area image.

[0063] Specifically, for the optimized segmentation results, the connected domain analysis process includes the following steps:

[0064] (1) Image scanning: Scan the optimized segmentation result, traversing each pixel from left to right and from top to bottom. When encountering a pixel with a value of 1 (stone area), start to check whether it has been marked as part of a connected domain. 0 represents a non-stone area and 1 represents a stone area, which is explained in the confidence map above and will not be explained in detail here.

[0065] (2) Marking connected domains: Use the depth-first search (DFS) or breadth-first search (BFS) algorithm to traverse the adjacent pixels around the pixel and determine all the pixels connected to it. All pixels connected to the pixel are marked as the same connected domain.

[0066] (3) Connected domain grouping: Assign a unique label (connected domain number) to each independent stone region. Whenever a new stone region is found, assign it a new number and continue scanning the image.

[0067] (4) Region screening: All detected connected regions are screened to remove possible small noise regions. Regions smaller than a certain area threshold are removed, as these regions are likely to be noise or small artifacts.

[0068] (5) Output of connected domain labeling map: For all the filtered connected regions, a labeling map or segmentation map is generated, in which each connected region is assigned a unique label. These labels correspond to different stone regions in the image.

[0069] Finally, a complete image of the stone area is formed.

[0070] It should be noted that the connected domain analysis method effectively eliminates the problem of breaks or omissions in the segmentation process by classifying areas with the same pixel values ​​into a connected area, ensuring that the stone area is completely extracted. This is especially important for dealing with cases where the stone has complex morphology and unclear boundaries. It can ensure that the stone area in the image is fully captured, thereby avoiding the risk of missed or mis-segmentation. The connected domain analysis method can refine the optimized segmentation results, especially when dealing with stones with complex morphology and different sizes, and can ensure that small stones and irregularly shaped stones can also be correctly identified. Through precise area division, it can avoid the impact of noise or artifacts on the segmentation quality, ensuring that the extracted stone area conforms to the actual anatomical characteristics.

[0071] See Figure 2 As shown, an embodiment of the present invention further provides an image segmentation system for stone removal, the system comprising:

[0072] The first module is configured to perform real-time image acquisition of the target part based on a pre-deployed stone image acquisition device to obtain an initial stone image, and pre-process the initial stone image to obtain a pre-processed initial stone image;

[0073] The second module is configured to input the preprocessed initial stone image into a deep learning model based on a convolutional neural network to obtain a preliminary segmentation result;

[0074] The third module is configured to optimize the preliminary segmentation result using the Sobel operator combined with the closing operation to obtain the optimized segmentation result;

[0075] The fourth module is configured to use the connected domain analysis method on the optimized segmentation results to obtain a complete stone area image.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. An image segmentation method for stone removal, characterized in that: include: Performing real-time image acquisition of the target area based on a pre-deployed stone image acquisition device to obtain an initial stone image, and preprocessing the initial stone image to obtain a preprocessed initial stone image; The preprocessed initial stone image is input into a deep learning model based on a convolutional neural network to obtain preliminary segmentation results; The preliminary segmentation results are optimized using the Sobel operator combined with the closing operation to obtain the optimized segmentation results; The connected domain analysis method is used on the optimized segmentation results to obtain a complete image of the stone area; The preprocessing of the initial stone image to obtain the preprocessed initial stone image specifically comprises: removing noise from the initial stone image using a multi-scale filtering method to obtain a denoised stone image; and performing contrast enhancement and histogram equalization on the denoised stone image to obtain a preprocessed initial stone image; The preprocessed initial stone image is input into a deep learning model based on a convolutional neural network to obtain a preliminary segmentation result. Specifically, the preprocessed initial stone image is input into a deep learning model based on a convolutional neural network, local edge information of the preprocessed initial stone image is extracted through shallow features in the encoder, global context information of the preprocessed initial stone image is extracted through deep features, and finally, the shallow features and the deep features are fused through a jump connection and decoded by a decoder to obtain a confidence map. A threshold is set for the confidence map, and an area in the confidence map that is larger than the set threshold is automatically extracted as a seed point. The preliminary segmentation result is obtained by combining the region growing method. The convolutional neural network-based deep learning model adopts an improved U-Net architecture, that is, a residual module, a parallel convolution module and an attention module are combined on the original U-Net architecture. The improved U-Net architecture includes an encoder and a decoder. The encoder includes multiple convolution modules and a pooling layer. Each convolution module in the multiple convolution modules is added with a residual module, and finally a parallel convolution module is added. The decoder includes a deconvolution module, a feature fusion module, a convolution module and a parallel convolution module. The feature fusion module includes an attention module. The preliminary segmentation result is optimized using the Sobel operator combined with a closing operation to obtain an optimized segmentation result. The specific content is: the Sobel operator is used to perform edge detection on the preliminary segmentation result, and the edge information of the preliminary segmentation result is extracted by calculating the gradient of the image to obtain the stone area. Then, an expansion operation is performed on the stone area to expand the stone area. Finally, an corrosion operation is performed to restore the original shape of the stone area and remove the noise introduced in the expansion operation, thereby finally forming an optimized segmentation result.

2. The image segmentation method for stone removal according to claim 1, characterized in that: The method further comprises extracting local edge information of the preprocessed initial stone image through shallow features in the encoder, specifically comprising: inputting the preprocessed initial stone image into the encoder, performing a convolution operation with a 3×3 convolution kernel after normalization, introducing nonlinear expression capability through an activation function, and then performing a convolution operation with multiple 3×3 convolution kernels in a convolution module to extract basic edge information, texture, and low-level features from the preprocessed initial stone image; normalizing the basic edge information, texture, and low-level features from the preprocessed initial stone image, performing a maximum pooling operation on the basic edge information, texture, and low-level features, and then performing splicing on the parallel convolution module to form local edge information. The basic edge information includes boundary lines and contour lines in the pre-processed initial stone image; The texture includes the roughness of the stone surface and the degree of concavity and convexity of the stone surface; The low-level features include grayscale contrast and color contrast between the stone and the pre-processed initial stone image background.

3. The image segmentation method for stone removal according to claim 2, characterized in that: The global context information of the preprocessed initial stone image is extracted through deep features. Specifically, the preprocessed initial stone image is input into the encoder, normalized and then subjected to a convolution operation with a 3×3 convolution kernel, nonlinear expression capability is introduced through an activation function, and then local features in the preprocessed initial stone image are extracted through multiple 3×3 convolution operations in multiple convolution modules. After each convolution operation, maximum pooling is performed through a pooling layer to aggregate the local features into global context information. The global context information obtained after each convolution operation is integrated through a parallel convolution module to obtain the global context information of the preprocessed initial stone image.

4. The image segmentation method for stone removal according to claim 3, characterized in that: Finally, the shallow features and deep features are fused through jump connections and decoded by the decoder to obtain a confidence map. The specific content is: the shallow features and deep features are input into the decoder, first passed through the deconvolution module, and the inverse convolution operation is performed, and then the jump connection is performed to fuse and splice the shallow features and deep features that have passed the deconvolution module to obtain a fused feature map, and then the fused feature map is convolved. At the same time, the attention module is combined to improve the accuracy of the fused feature map, and finally the fused feature map with improved accuracy is mapped to a single-channel confidence map through a 1×1 convolution layer.

5. The image segmentation method for stone removal according to claim 1, characterized in that: The connected domain analysis method is used on the optimized segmentation results to obtain a complete stone area image. The specific content is: the optimized segmentation results are divided into regions based on pixel values, and the same pixel values ​​are classified into the same stone area, thereby converting the optimized segmentation results into a complete stone area image.

6. An image segmentation system for stone removal, implemented as the image segmentation method for stone removal according to any one of claims 1 to 5, characterized in that: The system comprises: The first module is configured to perform real-time image acquisition of the target part based on a pre-deployed stone image acquisition device to obtain an initial stone image, and pre-process the initial stone image to obtain a pre-processed initial stone image; The second module is configured to input the preprocessed initial stone image into a deep learning model based on a convolutional neural network to obtain a preliminary segmentation result; The third module is configured to optimize the preliminary segmentation result using the Sobel operator combined with the closing operation to obtain the optimized segmentation result; The fourth module is configured to use the connected domain analysis method on the optimized segmentation results to obtain a complete stone area image.

Images