Spatial redundancy compression system based on spatial channel attention mechanism of neural network

Abstract

The present invention discloses a spatial redundancy compression model based on a spatial channel attention mechanism of a neural network, and specifically relates to the field of image processing technology. f 1 , f 2 The two layers of shallow features are fused. Then the fused shallow features and f 3 , f 4 , f 5 The three high-level features are aggregated to obtain a global map. f 2 , advanced features f 3 , f 4 , f 5 The four feature layers use spatial channel enhancement convolution modules to obtain multi-scale feature information. The obtained feature information is then used with the reverse gated channel transformation module to gradually mine the identification area, maintain global connections and effectively calculate. Then, a deep supervision strategy is used during the training process to supervise the output of the four feature layers and the global features to optimize the model performance. Finally, the output prediction image is obtained through the Sigmoid function.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a spatial redundancy compression system based on a spatial channel attention mechanism of a neural network. Background Art

[0002] Image segmentation is a basic technology in the field of computer vision. It is the process of dividing an image into multiple regions or objects. Image segmentation is particularly important in medicine because it can help doctors and researchers identify and quantify lesions, organs, and other biomarkers. It is commonly used in X-ray, CT, MRI, and ultrasound images, which place higher demands on segmentation algorithms due to their complexity and high noise characteristics. Medical image segmentation is widely used in clinical practice, such as classifying various organs in a given tomographic image, such as pancreas segmentation, multi-organ segmentation, etc., or screening for abnormal areas in a given area of ​​the body, such as brain tumor or polyp segmentation.

[0003] Polyps are small fleshy growths on the inner wall of the intestine. They may be benign, but sometimes they may develop into malignant tumors. Therefore, accurate polyp segmentation is crucial for the early detection and treatment of colorectal cancer. The gold standard for rectal cancer detection is colonoscopy, which can effectively prevent rectal cancer by screening and removing precancerous lesions. However, colonoscopy requires professional doctors to perform detection and segmentation, which is very time-consuming. However, manual colonoscopy is limited by the operator's status and experience, and an average of 26% of polyps are missed in video examinations. Therefore, using machine deep learning to assist clinicians in polyp detection and segmentation is an effective solution. Since the size, shape, color and texture of the polyps themselves are quite variable, and the boundary with the background intestinal wall is blurred, these complex situations make polyp segmentation very challenging. Therefore, a deep learning method that can detect rectal polyps early plays a vital role in the treatment of rectal cancer.

[0004] In recent years, with the development of deep learning technology, the field of medical image segmentation has made significant progress. Convolutional neural networks (CNNs) have become the model of choice for many medical image segmentation tasks because they can automatically learn image features and effectively perform regional segmentation. In particular, in polyp segmentation, researchers have used the powerful ability of CNNs to identify subtle features of polyps, improving the accuracy and efficiency of segmentation. In addition, researchers have explored a variety of data augmentation and regularization techniques to improve the generalization ability of the model, as well as the use of multimodal data fusion to combine the advantages of different imaging technologies. These methods help to improve the robustness of segmentation algorithms in different patients and under different imaging conditions.

[0005] Despite these advances, medical image segmentation, especially polyp segmentation, still faces many challenges. The boundaries of polyps exist in mucosa of different shapes, states, and colors, which leads to complex diversity of polyp boundaries. This diversity, coupled with downgraded decoding, poses challenges to the segmentation of detailed parts of the boundaries during the decoding process, resulting in blurred polyp boundaries. However, the variation in size and shape of polyps, inconsistent textures, and unclear boundaries often pose great challenges to polyp segmentation. In addition, in deep neural networks for polyp segmentation, there is often a lot of redundancy in the spatial and channel dimensions of feature maps. Summary of the invention

[0006] To address these problems, a spatial redundancy compression system based on the spatial channel attention mechanism of a neural network is proposed.

[0007] To this end, the present invention provides a spatial redundancy compression system based on a spatial channel attention mechanism of a neural network to solve the problems raised in the background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solutions: a spatial redundancy compression system based on a spatial channel attention mechanism of a neural network, the compression system comprising a pre-trained backbone network Res2Net-50, a shallow feature fusion module, an edge enhancement multi-scale feature fusion module, a spatial channel enhancement convolution module and a reverse gated channel transformation module;

[0009] Shallow feature fusion module (SFF), the backbone network Resnet-50 divides the polyp image input into five feature layers, and uses the spatial channel attention mechanism to convolve the shallow features f1 and f2 to obtain f1 ’ and f2 ’ , and then use the shallow feature fusion module to fuse the two to obtain the fused shallow feature sf1;

[0010] Edge-enhanced multi-scale feature fusion module, which combines high-level features {f i , i=3,4,5} three high-level feature layers and fused shallow features sf1, through the decoder PD, through parallel connection to aggregate all features, the decoder is composed of S g = PD(sf1,f3, f4, f5) to calculate and get the global map S g ; After deep supervision, the network is constrained at different scales to gain more attention on the boundaries; finally, the Sigmoid function is used to obtain an image with enhanced boundaries;

[0011] The spatial channel enhanced convolution module (SCEConv) is used to obtain multi-scale feature information for the shallow feature f2 and the high-level feature f3, f4, and f5. The obtained feature information is then used by the reverse gated channel transformation module (RGCT) to gradually mine the recognition area, maintain global connection and effectively calculate. A deep supervision strategy is used during the training process to supervise the side outputs of the four feature layers and the global features. The output prediction image is obtained through the Sigmoid function.

[0012] The spatial channel enhanced convolution module includes:

[0013] The SRU spatial reconstruction unit separates the feature map with rich information from the feature map with less information through a separation operation, and then adds these two parts of the feature map through a reconstruction operation to generate a feature map with rich information;

[0014] The CRU channel reconstruction unit segments, transforms and fuses the feature map with rich information in sequence to obtain the original feature map with reduced feature channels;

[0015] The CBAM convolutional block attention module includes a channel attention module and a spatial attention module. The channel attention module first performs global maximum pooling and global average pooling operations on each channel of the original feature map to capture the global characteristics of the channel; the pooled features are modeled through the fully connected layer and the activation function to obtain the channel attention weights suitable for applying to the feature map obtained by the CRU channel reconstruction unit; these attention weights are used to weight the channels of the original feature map to generate a weighted channel feature map; this step enhances the channel features that contribute to the task while reducing the impact of unimportant channels.

[0016] In the spatial attention module, for the original input feature map, the maximum pooling and average pooling are implemented along the channel direction to capture features of different scales, thereby obtaining contextual information; the two pooling results are spliced ​​in the channel dimension to form a feature map that integrates multi-scale contextual features; these fused features are refined using convolution operations to generate spatial attention weights; the Sigmoid function is applied to activate these spatial attention weights to ensure that the weight values ​​are in the range of 0 to 1; these spatial attention weights are multiplied by each pixel position of the original feature map to achieve weighted processing of the feature map. This process enhances the feature response of key areas in the image, while suppressing the features of those unimportant areas, thereby optimizing the expressiveness of the feature map. Finally, the weighted channel feature map and the weighted spatial feature map are multiplied element by element to obtain the attention-enhanced feature.

[0017] Preferably, the specific steps of the SRU spatial reconstruction unit generating a feature map with rich information include:

[0018] First, SRU uses trainable parameters in the group normalization layer to evaluate the spatial information of different feature maps;

[0019] Then, by performing sigmoid function and threshold processing on the trainable parameters, the weights are mapped to the range of (0,1) to obtain the informative weight W1 and the redundant weight W2;

[0020] Finally, the informative weight W1 and redundant weight W2 are multiplied by the input feature X to obtain the informative feature and redundant information features ; Through cross-reconstruction operation, The upper part and The lower part Performing element-by-element summation, we get , and then The upper part and The lower part Performing element-by-element summation, we get ; Finally, These two parts of features are fully combined to obtain a spatial fine feature map ;

[0021] The entire reconstruction operation is expressed as:

[0022] (1)

[0023] in, is element-wise multiplication, is the element-wise sum, For series connection.

[0024] Preferably, the input space is refined into feature maps It is divided into two parts, the upper part and the lower part; then, the upper part and the lower part are transformed respectively; the upper part uses efficient convolution operation to extract high-level representative information to obtain output Y1; the lower part generates a feature map with shallow hidden details through 1×1 convolution operation, and reuses the features of the original input, and takes the union of the two to obtain output Y2; finally, for the outputs Y1 and Y2, first use global average pooling to generate pooled features S1 and S2; then perform Softmax function on the pooled features S1 and S2 to obtain two sets of feature weight vectors β1 and β2, and finally use the feature weight vector to obtain the final output:

[0025]

[0026] Preferably, the output of the channel attention module Calculated by formula (3):

[0027] (3);

[0028] in, is the input feature map, is a multi-layer perceptron, and represents global average pooling and global maximum pooling operations, is the Sigmoid activation function.

[0029] Preferably, the output of the spatial attention module Calculated using the following formula:

[0030] (4);

[0031] in, Represents a 7x7 convolution operation, Indicates that the global average pooling and maximum pooling results are concatenated along the channel.

[0032] Preferably, assuming the input feature Represents the activation features in the convolutional network, and the reverse gated channel transformation module performs the following transformation on it:

[0033] GCT i (5)

[0034] Where C represents the number of channels, H represents the height, and W represents the width; GCT i It represents the gated channel transformation module. represents the trainable parameters, Helps to embed the adaptability of the output, Used to control the activation threshold and determine the behavior of the gated channel conversion module in each channel;

[0035] Side output S i Perform Sigmoid operation, and then use reverse operation to get the reverse output R i , where the reverse operation is:

[0036] R i =1 - Sigmoid(S i )(6)

[0037] Among them, 1 - Sigmoid (Si) is the reverse operation;

[0038] The overall reverse gated channel transformation formula is:

[0039] RGCT i = GCT i ⨀ R i (7)

[0040] Among them, RGCT i Represents the reverse gated channel transformation, and ⨀ means element-wise multiplication.

[0041] The present invention has the following advantages:

[0042] The image is input into the system of the present invention, and the backbone network Resnet-50 divides the network into 5 feature layers. First, the shallow feature fusion module (SFF) is used to fuse the two shallow features of f1 and f2. Then, the fused shallow features and the three high-level features of f3, f4, and f5 are aggregated through parallel partial decoders to obtain a global mapping map. Then, the spatial channel enhancement convolution module (SCEConv) is used for the four feature layers of f2, f3, f4, and f5 to obtain multi-scale feature information. Then, the obtained feature information is used using the reverse gated channel transformation module (RGCT) to gradually mine the identification area, maintain global connection and effectively calculate. Then, a deep supervision strategy is used in the training process to supervise the side outputs of the four feature layers and the global features to optimize the system performance. Finally, the output prediction image is obtained through the Sigmoid function, which solves the problem of blurred boundaries of polyp segmentation and a large amount of redundancy in the spatial and channel dimensions of feature mapping. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A1-A5 in the figure are pictures of serrated lesion polyps;

[0044] Figure 1 B1-B5 in the figure are pictures of adenomatous polyps;

[0045] Figure 2 Segmentation effect diagrams of the six networks provided in this embodiment, namely U-Net, U-Net++, SFA, PraNet, SANet and EMFFNet, on five public datasets;

[0046] Figure 3 The segmentation effect diagram of the three networks PraNet, SANet and EMFFNet provided in this embodiment on the private dataset;

[0047] Figure 4 Ablation experiment results of the EMFFNet system provided in this embodiment;

[0048] Figure 5 Schematic diagram of the compression system EMFFNet provided in this embodiment. DETAILED DESCRIPTION

[0049] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] The training set of this embodiment consists of 900 images of Kvasir-SEG and 550 images of CVC-ClinicDB. The images of the Kvasir and CVC-ClinicDB datasets are randomly divided into three parts, 80% of which are used for training, 10% for verification, and 10% for testing.

[0051] In order to verify the effectiveness of the system of the present invention, this embodiment conducted a comparative experiment on seven benchmark colonoscopy datasets. The specific situation of each dataset is as follows:

[0052] Kvasir: This dataset was collected by Vestre Viken Health Trust in Norway and contains 1,000 polyp images and their corresponding ground truths from colonoscopy video sequences. The ground truths were manually annotated by medical doctors and then verified by experienced gastroenterologists. The resolution of the images ranges from 332×487 to 1920×1072 pixels.

[0053] CVC-ClinicDB: This dataset contains 612 images collected from 29 colonoscopy video sequences with a resolution of 288 × 384. These images contain multiple examples of polyps, and masks corresponding to the areas covered by polyps in the images are provided to support image segmentation tasks.

[0054] CVC-ColonDB: This dataset consists of 380 polyp images and their corresponding ground truths with a resolution of 500 × 570. These images were generated from 15 different videos, and experts ensured that the extracted frames showed significantly different perspectives by excluding similar frames and manually annotated the 380 frames.

[0055] ETIS: This dataset contains 196 polyp images extracted from 34 colonoscopy videos with a resolution of 966 × 1225. The dataset includes variations in the size and shape of the coral polyps, blurred boundaries, and complex backgrounds, making the segmentation task extremely challenging.

[0056] CVC-300: This dataset is a test dataset from EndoScene, including 60 polyp images with a resolution of 500 × 574. This example selects CVC-300 as the test set, which contains a total of 60 samples for evaluation.

[0057] NICE_1 and NICE_2: These two data sets come from the A University Affiliated Hospital B and the C Traditional Chinese Medicine Hospital, respectively, and include 1459 intestinal polyp images, of which Dataset 1 contains 467 serrated lesion polyps and Dataset 2 contains 992 adenomatous polyps, which are used for external testing of the system. These data sets include images of different sizes, among which images of 1280x995 and 1504x1080 sizes are more common, accounting for 25.3% and 23.6% of the total, respectively. The external test set is specifically used to evaluate the performance of the system. It is not used in the training and parameter adjustment process to ensure the objectivity and independence of its evaluation results. In order to improve the adaptability and stability of the system to different situations, this embodiment selects endoscopic images containing different intestinal cleanliness, includes patients who underwent colonoscopy within the specified period, and all polyps found have been removed and histopathologically analyzed, excluding those with inflammatory bowel disease, active colitis, coagulopathy, familial polyposis, and those who undergo emergency colonoscopy.

[0058] In this embodiment, all polyps identified by colonoscopy were removed and subsequently subjected to histopathological examination. The results of the pathological diagnosis were fully recorded in the electronic medical record system. Except for those images showing no polyps, all images and videos recorded polyps were confirmed by pathology. If there was a discrepancy between the pathological results of the biopsy and resection specimens, this embodiment would give priority to the pathological results of the resection specimen.

[0059] like Figure 1 As shown, this embodiment classifies colonoscopic images according to pathological results, and divides colon polyps into two categories: adenomatous polyps (including tubular adenomas, villous adenomas, and tubulovillous adenomas) and serrated lesions (including hyperplastic polyps, sessile serrated lesions, and serrated adenomas). This classification is based on three main considerations: first, most colon polyps can be clearly classified into two categories: adenomatous or serrated lesions; second, pathologists have high consistency in distinguishing these two major types of lesions, but low consistency in distinguishing hyperplastic polyps and sessile serrated lesions; finally, the number of other types of polyps in the dataset of this embodiment is small, which is not enough to support the development of a robust multi-category system. To ensure the quality of the image, this embodiment uses a regularly maintained high-resolution endoscope and performs bowel preparation before colonoscopy according to the bowel preparation guidelines. All images included in the study have been anonymized to remove all information that may identify the patient.

[0060] In this embodiment, in order to comprehensively evaluate the performance of the compression system EMFFNet, a variety of quantitative evaluation indicators are used. First, this embodiment uses two basic indicators, average Dice and average IoU. In addition, four widely used measurement methods in the field of object detection are further introduced, including weighted Dice measurement , which corrects the problem of equal importance of all pixels in the traditional Dice metric; Mean absolute error (MAE), used to evaluate pixel-level accuracy, the lower the value, the better; Enhanced alignment metric , used to evaluate the similarity at the pixel level and the global level; and the structural metric Sα, used to measure the structural similarity between the predicted results and the true labels. This embodiment follows the existing state-of-the-art (SOTA) method and uses these six metrics - mDice, mIoU,, Sα, and MAE to evaluate the performance of the method in this embodiment and facilitate comparative analysis with other existing technologies. These metrics have been applied in a large number of literatures and have been proven to be effective in polyp segmentation tasks. Through these comprehensive evaluations, this embodiment can fully understand the performance of the system relative to the ground truth labels and make effective comparisons with other technologies. The evaluation toolbox comes from https: / / github.com / DengPingFan / PraNet.

[0061] In this example, the segmentation results are described by comparing the compression system EMFFNet with five state-of-the-art polyp segmentation methods. Based on the visual results, the results of the compression system EMFFNet of this example are closest to GroundTruth, and the compression system EMFFNet of this example outperforms the other compared methods in dealing with different challenging factors. Specifically, from Figure 2 As can be seen, in the first row, the background image in the polyp image is not clear and there are water droplet-shaped interferences, which increases the challenge of segmentation. In this case, U-Net, PraNet and SANet will produce some errors of over-segmentation areas, while SFA is completely unable to segment them. In the second and third rows, the polyps are almost fused with the background and are large in size, which makes it extremely difficult for the segmentation method to fully identify the boundaries of the polyps. Under this challenge, the system of this embodiment accurately segments the area. In the small-sized segmentation process in the fourth row, and in the segmentation of multiple small-sized polyps in the fifth row, the system of this embodiment still completes the task more accurately than other methods.

[0062] In this example, the datasets NICE_1 and NICE_2 (i.e. Figure 3 In Ours, Figure 3This embodiment compares several extremely challenging segmentations including small polyps, large polyps, and multiple polyps. The compression system EMFFNet of this embodiment still performs well.

[0063] Overall, the quantitative comparison further confirms that the system of this embodiment exhibits excellent performance in dealing with various challenging factors encountered during polyp segmentation.

[0064] After the compression system EMFFNet completes the training data set, it is first evaluated on the Kvasir and CVC-ClinicDB data sets. Table 1 shows the quantitative performance comparison of the EMFFNet method of this embodiment with other SOTA methods. The results show that the performance of the compression system EMFFNet is significantly better than other models. These models include the two most classic methods in the field of medical segmentation, U-Net and U-Net++, ResUNet++, a variant of ResNet and U-Net++, SFA, the originator of the selective feature aggregation network, and two advanced PraNet and SANet methods. As shown in Table 1, a comprehensive comparison of the system of this embodiment with past SOTA methods under six metrics is shown. All indicators of EMFFNet on the two datasets of Kvasir and CVC-ClinicDB have been improved. This is because the SCEConv module proposed in this embodiment can effectively extract feature information in the image considering the spatial and channel dimensions, and adaptively select and weight different parts of the feature map, thereby improving the system's ability to focus on the target area. In addition, the multi-scale edge enhancement feature fusion module proposed by the compression system EMFFNet can effectively enhance the edge information of the image, thereby capturing more and stronger edge features.

[0065]

[0066] In order to evaluate the generalization ability of the system, this embodiment selects CVC-300, ETIS, and CVC-ColonDB datasets for comparative experiments. Table 2 shows the qualitative results of the compression system EMFFNet proposed in this embodiment and other state-of-the-art models on five datasets.

[0067]

[0068] Since U-Net, U-Net++ and SFA have poor effects on NICE_1 and NICE_2 datasets, their data will not be collected for comparison in this article, and only PraNet and SANet mentioned in the article will be used. Table 3 is the qualitative results of EMFF-Net compared with 5 existing methods on the datasets NICE1 and NICE2 of this embodiment. It can be seen from the table that the dataset of this embodiment has a certain stability in production, and the three models have good performance on the dataset. Among them, the accuracy of the EMFFNet system of this embodiment on the NICE_1 and NICE_2 datasets is higher than that of PraNet and SANet.

[0069]

[0070] In this experiment, this example describes in detail the impact of each module on the performance of the entire system. In terms of evaluation methods, two common evaluation criteria in the field of medical image segmentation, mDice and mIoU, are adopted. This example uses PraNet as the baseline and evaluates the effectiveness of the three modules SFF, SCEConv, and RGCT proposed in this paper by replacing or deleting them. The test results are shown in Figure 2. Figure 4 As shown, the system of this embodiment is still effective.

[0071] SFF Validity:

[0072] In the system of this embodiment, the SFF module is designed to fuse and extract shallow features, which are incorporated into the segmentation network to provide boundary context information. In order to study the effectiveness of the SFF module, this embodiment performs an ablation study by removing it from the system of this embodiment, denoted as "w / o SFF". In addition, this embodiment adopts a simple and effective gate fusion strategy to integrate low-level features (i.e., f1 and f2). The experimental results of the ablation study are shown in Table 4. As shown in the table, by comparing "w / o SFF" with the system, it can be observed that the method using boundary-aware features can improve the segmentation performance.

[0073] RGCT effectiveness:

[0074] In the system of this embodiment, the RGCT module is designed to analyze and understand the local features of polyps, which are applied to the segmentation network to enhance the polyp boundary information. In order to study the effectiveness of the RGCT module, this embodiment performs an ablation study by removing it from the system of this embodiment, denoted as "w / o RGCT". The experimental results of the ablation study are shown in the table. As shown in Table 4, by comparing "w / o RGCT" with the system of this embodiment, it can be observed that the method of the reverse gated channel of this embodiment can improve the segmentation performance.

[0075] SCEConv Effectiveness:

[0076] In order to verify the effectiveness of SCEConv in the system of this embodiment, the boundary features are directly merged into the segmentation network using a cascade operation, which is represented as "w / o SCEConv". As shown in Table 4, the proposed SCEConv is effective in merging polyp edge features into the segmentation network, improving the performance of the segmentation network.

[0077]

[0078] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.

Claims

1. A spatial redundancy compression system based on the spatial channel attention mechanism of a neural network, characterized by: The compression system includes a pre-trained backbone network Res2Net-50, a shallow feature fusion module, an edge-enhanced multi-scale feature fusion module, a spatial channel-enhanced convolution module, and a reverse-gated channel transformation module; Shallow feature fusion module: The pre-trained backbone network Resnet-50 divides the polyp image input into five feature layers, and uses the spatial channel attention mechanism to convolve the shallow features f1 and f2 to obtain f1 ’ and f2 ’ , and then use the shallow feature fusion module to convolute f1 ’ and f2 ’ The fusion is used to obtain the fused shallow feature sf1; Edge-enhanced multi-scale feature fusion module, which combines high-level features {f i , i=3,4,5} three high-level feature layers and fused shallow features sf1, through the decoder PD, through parallel connection to aggregate all features, the decoder is composed of S g = PD(sf1, f3,f4, f5) to calculate and get the global map S g ; After deep supervision, the network is constrained at different scales to gain more attention on the boundaries; finally, the Sigmoid function is used to obtain an image with enhanced boundaries; The spatial channel enhanced convolution module is used to obtain multi-scale feature information for the shallow feature f2 and the high-level feature f3, f4, and f5. The obtained feature information is then used in the reverse gated channel transformation module to gradually mine the identification area, maintain global connection and effectively calculate. A deep supervision strategy is used in the training process to supervise the side outputs of the four feature layers and the global features. The output prediction image is obtained through the Sigmoid function. The spatial channel enhanced convolution module includes: The SRU spatial reconstruction unit separates the feature map with rich information from the feature map with less information through a separation operation, and then adds these two parts of the feature map through a reconstruction operation to generate a feature map with rich information; The CRU channel reconstruction unit sequentially segments, transforms and fuses the feature map with rich information to obtain a feature map with reduced feature channels; The CBAM convolutional block attention module includes a channel attention module and a spatial attention module. The channel attention module first performs global maximum pooling and global average pooling operations on each channel of the original feature map to capture the global features of the channel; the pooled features are modeled through the fully connected layer and the activation function to obtain the channel attention weights suitable for the feature map obtained by the CRU channel reconstruction unit; these attention weights are used to weight the channels of the original feature map to generate a weighted channel feature map; the spatial attention module captures features of different scales for the input original feature map by implementing maximum pooling and average pooling along the channel direction to obtain context information; the features of the maximum pooling and average pooling are spliced ​​along the channel dimension to form a feature map that integrates multi-scale context features; the convolution operation is used to refine the fused features to generate spatial attention weights; the sigmoid function is applied to activate the spatial attention weights; the spatial attention weights are multiplied by each pixel position of the original feature map to achieve weighted processing of the spatial feature map, and finally the weighted channel feature map and the weighted spatial feature map are multiplied element by element to obtain the attention enhanced feature.

2. The spatial redundancy compression system based on the spatial channel attention mechanism of the neural network according to claim 1, characterized in that: The specific steps of the SRU spatial reconstruction unit to generate a feature map with rich information include: First, SRU uses trainable parameters in the group normalization layer to evaluate the spatial information of different feature maps; Then, by performing Sigmoid function and threshold processing on the trainable parameters, the weights are mapped to the range of (0,1) to obtain the informative weight W1 and the redundant weight W2; Finally, the informative weight W1 and redundant weight W2 are multiplied by the input feature X to obtain the informative feature and redundant information features ; Through cross-reconstruction operation, The upper part and The lower part Performing element-by-element summation, we get , and then The upper part and The lower part Performing element-by-element summation, we get ; Finally, The two parts of features are combined to obtain a spatially refined feature map ; The entire reconstruction operation is expressed as: in, is element-wise multiplication, is the element-wise sum, For series connection.

3. The spatial redundancy compression system based on the spatial channel attention mechanism of the neural network according to claim 2, characterized in that: Map the input spatial fine feature It is divided into two parts, namely the upper part and the lower part; then, the upper part and the lower part are transformed respectively; the upper part uses convolution operation to extract high-level representative information to obtain output Y1; the lower part generates a feature map with shallow hidden details through 1×1 convolution operation, and reuses the features of the original input, and takes the union of the two to obtain output Y2; finally, for the outputs Y1 and Y2, first use global average pooling to generate pooled features S1 and S2; then perform Softmax function on the pooled features S1 and S2 to obtain two sets of feature weight vectors β1 and β2, and finally use the feature weight vector to obtain the final output:

4. The spatial redundancy compression system based on the spatial channel attention mechanism of a neural network according to claim 1, characterized in that: Output of the channel attention module Calculated by formula (3): （3）； in, is the input feature map, is a multi-layer perceptron, and represents global average pooling and global maximum pooling operations, is the Sigmoid activation function.

5. The spatial redundancy compression system based on the spatial channel attention mechanism of the neural network according to claim 4, characterized in that: Output of the spatial attention module Calculated using the following formula: （4）； in, Represents a 7x7 convolution operation, Indicates that the global average pooling and maximum pooling results are concatenated along the channel.

6. The spatial redundancy compression system based on the spatial channel attention mechanism of a neural network according to claim 1, characterized in that: Assuming the input features Represents the activation features in the convolutional network, and the reverse gated channel transformation module performs the following transformation on it: GCT i (5) Where C represents the number of channels, H represents the height, and W represents the width; GCT i It represents the gated channel transformation module. represents the trainable parameters, Helps to embed the adaptability of the output, Used to control the activation threshold and determine the behavior of the gated channel conversion module in each channel; Side output S i Perform Sigmoid operation, and then use reverse operation to get the reverse output R i , where the reverse operation is: R i =1 - Sigmoid(S i )(6) Among them, 1 - Sigmoid (Si) is the reverse operation; The overall reverse gated channel transformation formula is: RGCT i = GCT i ⨀ R i (7) Among them, RGCT i Represents the reverse gated channel transformation, and ⨀ means element-wise multiplication.

Images