Information In Different Regions Research Articles

Towards a Flexible Semantic Guided Model for Single Image Enhancement and Restoration.

Low-light image enhancement (LLIE) investigates how to improve the brightness of an image captured in illumination-insufficient environments. The majority of existing methods enhance low-light images in a global and uniform manner, without taking into account the semantic information of different regions. Consequently, a network may easily deviate from the original color of local regions. To address this issue, we propose a semantic-aware knowledge-guided framework (SKF) that can assist a low-light enhancement model in learning rich and diverse priors encapsulated in a semantic segmentation model. We concentrate on incorporating semantic knowledge from three key aspects: a semantic-aware embedding module that adaptively integrates semantic priors in feature representation space, a semantic-guided color histogram loss that preserves color consistency of various instances, and a semantic-guided adversarial loss that produces more natural textures by semantic priors. Our SKF is appealing in acting as a general framework in the LLIE task. We further present a refined framework SKF++ with two new techniques: (a) Extra convolutional branch for intra-class illumination and color recovery through extracting local information and (b) Equalization-based histogram transformation for contrast enhancement and high dynamic range adjustment. Extensive experiments on various benchmarks of LLIE task and other image processing tasks show that models equipped with the SKF/SKF++ significantly outperform the baselines and our SKF/SKF++ generalizes to different models and scenes well. Besides, the potential benefits of our method in face detection and semantic segmentation in low-light conditions are discussed. The code and pre-trained models have been publicly available at https://github.com/langmanbusi/Semantic-Aware-Low-Light-Image-Enhancement.

(HTBNet)Arbitrary Shape Scene Text Detection with Binarization of Hyperbolic Tangent and Cross-Entropy.

The existing segmentation-based scene text detection methods mostly need complicated post-processing, and the post-processing operation is separated from the training process, which greatly reduces the detection performance. The previous method, DBNet, successfully simplified post-processing and integrated post-processing into a segmentation network. However, the training process of the model took a long time for 1200 epochs and the sensitivity to texts of various scales was lacking, leading to some text instances being missed. Considering the above two problems, we design the text detection Network with Binarization of Hyperbolic Tangent (HTBNet). First of all, we propose the Binarization of Hyperbolic Tangent (HTB), optimized along with which the segmentation network can expedite the initial convergent speed by reducing the number of epochs from 1200 to 600. Because features of different channels in the same scale feature map focus on the information of different regions in the image, to better represent the important features of all objects in the image, we devise the Multi-Scale Channel Attention (MSCA). Meanwhile, considering that multi-scale objects in the image cannot be simultaneously detected, we propose a novel module named Fused Module with Channel and Spatial (FMCS), which can fuse the multi-scale feature maps from channel and spatial dimensions. Finally, we adopt cross-entropy as the loss function, which measures the difference between predicted values and ground truths. The experimental results show that HTBNet, compared with lightweight models, has achieved competitive performance and speed on Total-Text (F-measure:86.0%, FPS:30) and MSRA-TD500 (F-measure:87.5%, FPS:30).

New insights in the 3-D rheological properties and collagen fibers orientation in murine periodontal ligaments

The rheological properties of the periodontal ligament are key parameters to understand the homeostatic stability of the tooth supporting apparatus. The objective of this study is to lay new insights on the rheological properties and structural information of different regions in murine periodontal ligament by using atomic force (AFM) and multi-photon microscopy (MPM). A significant variation in elasticity of different regions was measured. The elasticity of periodontal ligament showed a significant tendency to become softer towards the furcation, the nearest area to the center of resistance of the tooth. This can open a new prospective for connecting the rheological adaptation of periodontal ligament to the tooth geometry that defines the center of resistance. Another important finding revealed by the second harmonic generation (SHG) signal exhibited by collagen fibers and measured by MPM is that the orientation of fibers in the furcation region can both provide space for tooth vertical movement with high compressive loads and prevent horizontal tooth movement. Additionally, it was found that the dispersion of the angles at different levels of cutting indicates homogeneity in the directionality of fibers across different regions. These results provide an accurate description of the rheological properties and structural information of periodontal ligament, which can serve as a base for comparison with other local and systemic diseases that may influence the periodontal ligament.

Ultra short term power load forecasting based on the fusion of Seq2Seq BiLSTM and multi head attention mechanism.

Ultra-short-term power load forecasting is beneficial to improve the economic efficiency of power systems and ensure the safe and stable operation of power grids. As the volatility and randomness of loads in power systems, make it difficult to achieve accurate and reliable power load forecasting, a sequence-to-sequence based learning framework is proposed to learn feature information in different dimensions synchronously. Convolutional Neural Networks(CNN) Combined with Bidirectional Long Short Term Memory(BiLSTM) Networks is constructed in the encoder to extract the correlated timing features embedded in external factors affecting power loads. The parallel BiLSTM network is constructed in the decoder to mine the power load timing information in different regions separately. The multi-headed attention mechanism is introduced to fuse the BiLSTM hidden layer state information in different components to further highlight the key information representation. The load forecastion results in different regions are output through the fully connected layer. The model proposed in this paper has the advantage of high forecastion accuracy through the example analysis of real power load data.

A graph convolutional network with dynamic weight fusion of multi-scale local features for diabetic retinopathy grading

Diabetic retinopathy (DR) is a serious ocular complication that can pose a serious risk to a patient’s vision and overall health. Currently, the automatic grading of DR is mainly using deep learning techniques. However, the lesion information in DR images is complex, variable in shape and size, and randomly distributed in the images, which leads to some shortcomings of the current research methods, i.e., it is difficult to effectively extract the information of these various features, and it is difficult to establish the connection between the lesion information in different regions. To address these shortcomings, we design a multi-scale dynamic fusion (MSDF) module and combine it with graph convolution operations to propose a multi-scale dynamic graph convolutional network (MDGNet) in this paper. MDGNet firstly uses convolution kernels with different sizes to extract features with different shapes and sizes in the lesion regions, and then automatically learns the corresponding weights for feature fusion according to the contribution of different features to model grading. Finally, the graph convolution operation is used to link the lesion features in different regions. As a result, our proposed method can effectively combine local and global features, which is beneficial for the correct DR grading. We evaluate the effectiveness of method on two publicly available datasets, namely APTOS and DDR. Extensive experiments demonstrate that our proposed MDGNet achieves the best grading results on APTOS and DDR, and is more accurate and diverse for the extraction of lesion information.

RDPGNet: A road extraction network with dual-view information perception based on GCN

Automatic road extraction from satellite images is a popular research hotspot topic in the field of remote sensing. However, the complexity of road features and structures hinders their precise extraction. Road features have different saliency due to the occlusion of other objects and diverse road types. Road structures are scattered across various locations and affected by geographical conditions, population distribution, and transportation demands. These issues render current methods based on convolutional neural networks prone to fragmented and incomplete road extraction results. To address these issues, a Road Extraction Network with Dual-View Information Perception Based on GCN (RDPGNet) is proposed, where the GCN-based dual-view perceptron (GDVP) uses the superior information interaction capability of the GCN to explore road information under a dual view. In the GDVP, we first design the road feature saliency graph reasoning module (RFSG) and road structure homogeneous space graph reasoning module (RSHS) based on the features and structures. The RFSG obtains the similarity of road feature information in different regions as a similarity matrix to participate in the graph reasoning process, ensuring that equal treatment is given to regions with varying saliency levels. The RSHS projects the road structure onto a corresponding homogeneous space while aggregating and interacting with information through graph convolution, thereby enhancing the network’s perception of roads in diverse locations. Second, because multi-view information usually shares a potential common representation, a multi-view information fusion and alignment strategy (MVFA) is designed to comprehensively model road information. Experimental results obtained from two public datasets indicate that RDPGNet outperforms other state-of-the-art networks.

A lightweight convolution neural network based on joint features for Remote Sensing scene image classification

ABSTRACT Unlike natural images, remote sensing scene images usually contain one scene label and many object labels, and many object labels are arranged dispersedly, which brings great difficulties to feature extraction of scene label. To accurately identify scene labels from remote sensing scene images with multiple object labels, it is important to fully understand the global context of the image. In order to solve the challenges of multi-label scene images and improve the classification performance, a global context feature extraction module is proposed in this paper. The module combines the semantics information of different regions through a global pooling and three different scale sub-regions pooling, which makes the module have stronger ability of global feature representation. In addition, in order to fully understand the semantic content of remote sensing images, a three branch joint feature extraction module is constructed, which consists of the global context feature module, 3 × 3 convolution branch and identity branch are fused. Finally, a lightweight convolution neural network based on joint features (LCNN-JF) is constructed using traditional convolution, depthwise separable convolution, joint feature extraction module and classifier for remote sensing scene image classification. A series of experimental results on four datasets, UCM, AID, RSSCN and NWPU, demonstrate that the proposed method has better feature representation ability and can achieve better classification of remote sensing scene images.

Assessing geographical origin of Diqing wines based on their elemental and isotopic profiles

Characterization of grape and wine elemental constitution is essential in tracing their geographical origin. The traceability of wine provenance in China is conducive to implementing regional product quality protection. The aim of the present study was to investigate the multi-element profile of subregions and combine the variables of grapes and wine to assess the geographical origin. Grape and wine samples were collected from three towns in Deqin County, plateau of Diqing, China. The oxygen and carbon stable isotope ratios and the concentrations of 45 elements were measured. The interannual variability of element contents in grape and wine samples was less than 40 %. The correlation of elements between grape and wine samples can form fingerprint information of different regions. No significant variability in stable isotopes was observed among micro-vinifications from different wine-producing of the county. The grape data classification accuracy was 88.9 % by stepwise discriminant analysis (DA), and wine data classification accuracy was 95.2 % by an artificial neural network (ANN) model. The conclusion was that multielemental and isotopic information of grapes and wine will help to discriminate wine origin at the county level in China.

Matching medical staff to long term care facilities to respond to COVID-19 outbreak

BackgroundStaff shortage is a long-standing issue in long term care facilities (LTCFs) that worsened with the COVID-19 outbreak. Different states in the US have employed various tools to alleviate this issue in LTCFs. We describe the actions taken by the Commonwealth of Massachusetts to assist LTCFs in addressing the staff shortage issue and their outcomes. Therefore, the main question of this study is how to create a central mechanism to allocate severely limited medical staff to healthcare centers during emergencies.MethodsFor the Commonwealth of Massachusetts, we developed a mathematical programming model to match severely limited available staff with LTCF demand requests submitted through a designed portal. To find feasible matches and prioritize facility needs, we incorporated restrictions and preferences for both sides. For staff, we considered maximum mileage they are willing to travel, available by date, and short- or long-term work preferences. For LTCFs, we considered their demand quantities for different positions and the level of urgency for their demand. As a secondary goal of this study, by using the feedback entries data received from the LTCFs on their matches, we developed statistical models to determine the most salient features that induced the LTCFs to submit feedback.ResultsWe used the developed portal to complete about 150 matching sessions in 14 months to match staff to LTCFs in Massachusetts. LTCFs provided feedback for 2,542 matches including 2,064 intentions to hire the matched staff during this time. Further analysis indicated that nursing homes and facilities that entered higher levels of demand to the portal were more likely to provide feedback on the matches and facilities that were prioritized in the matching process due to whole facility testing or low staffing levels were less likely to do so. On the staffing side, matches that involved more experienced staff and staff who can work afternoons, evenings, and overnight were more likely to generate feedback from the facility that they were matched to.ConclusionDeveloping a central matching framework to match medical staff to LTCFs at the time of a public health emergency could be an efficient tool for responding to staffing shortages. Such central approaches that help allocate a severely limited resource efficiently during a public emergency can be developed and used for different resource types, as well as provide crucial demand and supply information in different regions and/or demographics.

Region-adaptive Concept Aggregation for Few-shot Visual Recognition

Few-shot learning (FSL) aims to learn novel concepts from very limited examples. However, most FSL methods suffer from the issue of lacking robustness in concept learning. Specifically, existing FSL methods usually ignore the diversity of region contents that may contain concept-irrelevant information such as the background, which would introduce bias/noise and degrade the performance of conceptual representation learning. To address the above-mentioned issue, we propose a novel metric-based FSL method termed region-adaptive concept aggregation network or RCA-Net. Specifically, we devise a region-adaptive concept aggregator (RCA) to model the relationships of different regions and capture the conceptual information in different regions, which are then integrated in a weighted average manner to obtain the conceptual representation. Consequently, robust concept learning can be achieved by focusing more on the concept-relevant information and less on the conceptual-irrelevant information. We perform extensive experiments on three popular visual recognition benchmarks to demonstrate the superiority of RCA-Net for robust few-shot learning. In particular, on the Caltech-UCSD Birds-200-2011 (CUB200) dataset, the proposed RCA-Net significantly improves 1-shot accuracy from 74.76% to 78.03% and 5-shot accuracy from 86.84% to 89.83% compared with the most competitive counterpart.

Spatial and temporal landslide distributions using global and open landslide databases

Landslide databases are a potential tool for the analysis of landslide susceptibility, hazard, and risk. Additionally, the spatio-temporal distribution of landslides and their correlation with their triggering factors are inputs that facilitate the evaluation of landslide prediction models and the determination of thresholds necessary for early warning systems (EWS). This study presents an analysis of four widely known global databases—the International Disaster database (EM-DAT), the Disaster Inventory System (DesInventar), the Global Landslide Catalog (GLC), and the Global Fatal Landslide database (GFLD)—which contain relevant landslide information for different regions of the world. These databases were analysed and compared by means of the spatio-temporal distributions of their records. Subsequently, these databases were merged and depurated to obtain a more robust database, namely the Unified Global Landslide Database (UGLD), with 161 countries, 37,946 landslides, and 185,753 fatalities registered between 1903 and 2020. The merging process among the databases resulted in a small number of repeated landslides, indicating that the databases collect very different landslide information and complement each other. Finally, an update of the spatial and temporal analysis of landslides in the world was performed with the new database, in which patterns, trends, and the main triggers were presented and analysed. The results obtained from the analysis of the UGLD database show the American and Asian continents as the continents with the highest number of landslides and associated fatalities, showing a bimodal and unimodal annual temporal pattern, respectively. Regarding the most frequent triggers of landslides, rainfall, anthropogenic intervention, and earthquakes stand out.

Temporal pain processing in the primary somatosensory cortex and anterior cingulate cortex

Pain is known to have sensory and affective components. The sensory pain component is encoded by neurons in the primary somatosensory cortex (S1), whereas the emotional or affective pain experience is in large part processed by neural activities in the anterior cingulate cortex (ACC). The timing of how a mechanical or thermal noxious stimulus triggers activation of peripheral pain fibers is well-known. However, the temporal processing of nociceptive inputs in the cortex remains little studied. Here, we took two approaches to examine how nociceptive inputs are processed by the S1 and ACC. We simultaneously recorded local field potentials in both regions, during the application of a brain-computer interface (BCI). First, we compared event related potentials in the S1 and ACC. Next, we used an algorithmic pain decoder enabled by machine-learning to detect the onset of pain which was used during the implementation of the BCI to automatically treat pain. We found that whereas mechanical pain triggered neural activity changes first in the S1, the S1 and ACC processed thermal pain with a reasonably similar time course. These results indicate that the temporal processing of nociceptive information in different regions of the cortex is likely important for the overall pain experience.

Study on the Identification and Detection of Walnut Quality Based on Terahertz Imaging.

Walnuts have rich nutritional value and are favored by the majority of consumers. As walnuts are shelled nuts, they are prone to suffer from defects such as mildew during storage. The fullness and mildew of the fruit impose effects on the quality of the walnuts. Therefore, it is of great economic significance to carry out a study on the rapid, non-destructive detection of walnut quality. Terahertz spectroscopy, with wavelengths between infrared and electromagnetic waves, has unique detection advantages. In this paper, the rapid and nondestructive detection of walnut mildew and fullness based on terahertz spectroscopy is carried out using the emerging terahertz transmission spectroscopy imaging technology. First, the normal walnuts and mildewed walnuts are identified and analyzed. At the same time, the image processing is carried out on the physical samples with different kernel sizes to calculate the fullness of the walnut kernels. The THz image of the walnuts is collected to extract the spectral information in different regions of interest. Four kinds of time domain signals in different regions of interest are extracted, and three qualitative discrimination models are established, including the support vector machine (SVM), random forest (RF), and k-nearest neighbor (KNN) algorithms. In addition, in order to realize the visual expression of walnut fullness, the terahertz images of the walnut are segmented with a binarization threshold, and the walnut fullness is calculated by the proportion of the shell and kernel pixels. In the frequency domain signal, the amplitude intensity from high to low is the mildew sample, walnut kernel, and walnut shell, and the distinction between walnut kernel, shell samples, and mildew samples is high. The overall identification accuracy of the aforementioned three models is 90.83%, 97.38%, and 97.87%, respectively. Among them, KNN has the best qualitative discrimination effect. In a single category, the recognition accuracy of the model for the walnut kernel, walnut shell, mildew sample, and reference group (background) reaches 94%, 100%, 97.43%, and 100%, respectively. The terahertz transmission images of the five categories of walnut samples with different kernel sizes are processed to visualize the detection of kernel fullness inside walnuts, and the errors are less than 5% compared to the actual fullness of walnuts. This study illustrates that terahertz spectroscopy detection can achieve the detection of walnut mildew, and terahertz imaging technology can realize the visual expression and fullness calculation of walnut kernels. Terahertz spectroscopy and imaging provides a non-destructive detection method for walnut quality, which can provide a reference for the quality detection of other dried nuts with shells, thus having significant practical value.

Research on the Key Technology of Web Data Extraction and Mining Based on the Probability Distribution

In the development of internet technology innovation, the quantity of big data information continues to rise, showing the characteristics of dynamic, heterogeneous, massive, and so on. How to explore valuable and potential knowledge from the network system is the main topic of research and scholars. Based on the understanding of web data extraction and web mining technology, this paper makes a systematic study of web text mining, proposes the most commonly used hidden Markov model in probability distribution, and constructs the corresponding text mining method. The hidden Markov model is a statistical model, which is mainly used to represent a Markov process with unknown parameters. The difficulty in practical application lies in how to define the hidden parameters of the process from observable parameters and then use these parameters to conduct in-depth research. Finally, the practical results show that the hidden Markov model can not only obtain the information of different regions but also deeply analyze the data set, which proves the feasibility of this research technology.

Lightweight Image Super-Resolution Based on Local Interaction of Multi-Scale Features and Global Fusion

In recent years, computer vision technology has been widely applied in various fields, making super-resolution (SR), a low-level visual task, a research hotspot. Although deep convolutional neural network has made good progress in the field of single-image super-resolution (SISR), its adaptability to real-time interactive devices that require fast response is poor due to the excessive amount of network model parameters, the long inference image time, and the complex training model. To solve this problem, we propose a lightweight image reconstruction network (MSFN) for multi-scale feature local interaction based on global connection of the local feature channel. Then, we develop a multi-scale feature interaction block (FIB) in MSFN to fully extract spatial information of different regions of the original image by using convolution layers of different scales. On this basis, we use the channel stripping operation to compress the model, and reduce the number of model parameters as much as possible on the premise of ensuring the reconstructed image quality. Finally, we test the proposed MSFN model with the benchmark datasets. The experimental results show that the MSFN model is better than the other state-of-the-art SR methods in reconstruction effect, computational complexity, and inference time.

Siamese Implicit Region Proposal Network With Compound Attention for Visual Tracking.

Recently, siamese-based trackers have achieved significant successes. However, those trackers are restricted by the difficulty of learning consistent feature representation with the object. To address the above challenge, this paper proposes a novel siamese implicit region proposal network with compound attention for visual tracking. First, an implicit region proposal (IRP) module is designed by combining a novel pixel-wise correlation method. This module can aggregate feature information of different regions that are similar to the pre-defined anchor boxes in Region Proposal Network. To this end, the adaptive feature receptive fields then can be obtained by linear fusion of features from different regions. Second, a compound attention module including a channel and non-local attention is raised to assist the IRP module to perform a better perception of the scale and shape of the object. The channel attention is applied for mining the discriminative information of the object to handle the background clutters of the template, while non-local attention is trained to aggregate the contextual information to learn the semantic range of the object. Finally, experimental results demonstrate that the proposed tracker achieves state-of-the-art performance on six challenging benchmark tests, including VOT-2018, VOT-2019, OTB-100, GOT-10k, LaSOT, and TrackingNet. Further, our obtained results demonstrate that the proposed approach can be run at an average speed of 72 FPS in real time.

Recognition of Taxi Violations Based on Semantic Segmentation of PSPNet and Improved YOLOv3

Taxi has the characteristics of strong mobility and wide dispersion, which makes it difficult for relevant law enforcement officers to make accurate judgment on their illegal acts quickly and accurately. With the investment of intelligent transportation system, image analysis technology has become a new method to determine the illegal behavior of taxis, but the current image analysis method is still difficult to support the detection of illegal behavior of taxis in the actual complex image scene. To solve this problem, this study proposed a method of taxi violation recognition based on semantic segmentation of PSPNet and improved YOLOv3. (1) Based on YOLOv3, the proposed method introduces spatial pyramid pooling (SPP) for taxi recognition, which can convert vehicle feature images with different resolutions into feature vectors with the same dimension as the full connection layer and solve the problem of repeated extraction of YOLOv3 vehicle image features. (2) This method can recognize two different violations of taxi (blocking license plate and illegal parking) rather than only one. (3) Based on PSPNet semantic segmentation network, a taxi illegal parking detection method is proposed. This method can collect the global information of road condition images and aggregate the image information of different regions, so as to improve the ability to obtain the global information orderly and improve the accuracy of taxi illegal parking detection. The experimental results show that the proposed method has excellent recognition performance for the detection rate of license plate occlusion behavior DR is 85.3%, and the detection rate of taxi illegal parking phenomenon DR is 96.1%.

Fine‐grained semantic ethnic costume high‐resolution image colorization with conditional GAN

Grayscale image colorization, especially for ethnic costume images, is highly challenging due to its rich and complex color features. The existing image colorization methods usually take the costume image as a whole in practical applications that lead to the ignorance of the semantic information of different parts of the costume. It is known that each part's color distribution of the ethnic costume is different. So, the color mapping of other parts is also diverse, which is determined by distinctive ethnic characteristics. This study introduces fine-grained level semantic information and proposes a high-resolution image colorization model for ethnic costumes targeting enhancement, inspired by semantic-level colorization. The semantic information of different regions of ethnic costumes has a significant impact on the performance of the coloring task. Using Pix2PixHD as the backbone network, we create a new network architecture that maintains color distribution correspondence and spatial consistency of costume images using fine-grained semantic information. In our network, we take the splice result of fine-grained semantic for ethnic costume and grayscale image as the conditions and then feed them into the generative adversarial networks. We also discuss and analyze the influences of the grayscale channel and fine-grained semantics on discriminator. Extensive experiments demonstrate that our method performs well compared with other state-of-the-art automatic colorization methods.

Research on Simulation and Test of Spatial Isolation Based on Geographic Information

In the process of site selection for the deployment of distributed systems, in order to solve the problem of inaccurate spatial isolation simulation results caused by the large influence of complex geographic information such as terrain and weather on the propagation of radio waves, and to solve the problem of inefficient and inefficient spatial isolation testing. This paper proposes to select suitable propagation models for different geographical information in different regions, clarifies the simulation calculation method, and proposes a test method to accurately determine the direction of the transceiver antenna for the minimum spatial isolation. By comparing and analyzing the simulation calculation results and test verification results, correction parameters such as simulation system error, atmospheric absorption loss, and geological absorption loss are proposed. This article analyzes and studies the correction value of radio wave transmission loss of S-band electromagnetic signal under different geographic information conditions such as snowfall and iron ore through an example. It can provide reference values for some of the influencing factors of electromagnetic signal propagation loss under geographic information conditions, and it can improve the accuracy of the simulation calculation results of the spatial isolation under the conditions of geographic information.

UNet retinal blood vessel segmentation algorithm based on improved pyramid pooling method and attention mechanism

The segmentation results of retinal vessels have a significant impact on the automatic diagnosis of retinal diabetes, hypertension, cardiovascular and cerebrovascular diseases and other ophthalmic diseases. In order to improve the performance of blood vessels segmentation, a pyramid scene parseing U-Net segmentation algorithm based on attention mechanism was proposed. The modified PSP-Net pyramid pooling module is introduced on the basis of U-Net network, which aggregates the context information of different regions so as to improve the ability of obtaining global information. At the same time, attention mechanism was introduced in the skip connection part of U-Net network, which makes the integration of low-level features and high-level semantic features more efficient and reduces the loss of feature information through nonlinear connection mode. The sensitivity, specificity, accuracy and AUC of DRIVE and CHASE_DB1 data sets are 0.7814, 0.9810, 0.9556, 0.9780; 0.8195, 0.9727, 0.9590, 0.9784. Experimental results show that the PSP-UNet segmentation algorithm based on the attention mechanism enhances the detection ability of blood vessel pixels, suppresses the interference of irrelevant information and improves the network segmentation performance, which is superior to U-Net algorithm and some mainstream retinal vascular segmentation algorithms at present.