Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874008
Xin Tian, Bomeng Li, Xiaodong Cheng, Xiangyang Shi
Accurate and effective behavior recognition of cows is the basis for realizing informationization, high efficiency and scale of animal husbandry farming. To address the limitations of traditional non-contact and contact for obtaining animal behavior information, this paper investigates the target detection based on YOLOv5 algorithm and the cow standing behavior recognition method for video analysis. This paper first introduces the target detection algorithm, then describes the target detection network model (YOLOv5Net), which extracts the relevant features of cow images and performs image target detection through training to recognize the standing behavior of cows in real time. To achieve effective recognition of cow standing and efficient extraction of cow targets in complex natural environments, the YOLOv5 model for cow standing recognition is explored[8]; finally, the implemented YOLOv5 model is evaluated and analyzed for environment modeling and target detection algorithm objectives, and the experimental results show that the experimental detection correctness accuracy is 97.6%, and the preprocessing time in detecting a single image is It can quickly and accurately identify the standing behavior of cows, which lays the foundation for basic behavior identification and localization of cows.
{"title":"Target detection and cow standing behavior recognition based on YOLOv5 algorithm","authors":"Xin Tian, Bomeng Li, Xiaodong Cheng, Xiangyang Shi","doi":"10.1109/ISPDS56360.2022.9874008","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874008","url":null,"abstract":"Accurate and effective behavior recognition of cows is the basis for realizing informationization, high efficiency and scale of animal husbandry farming. To address the limitations of traditional non-contact and contact for obtaining animal behavior information, this paper investigates the target detection based on YOLOv5 algorithm and the cow standing behavior recognition method for video analysis. This paper first introduces the target detection algorithm, then describes the target detection network model (YOLOv5Net), which extracts the relevant features of cow images and performs image target detection through training to recognize the standing behavior of cows in real time. To achieve effective recognition of cow standing and efficient extraction of cow targets in complex natural environments, the YOLOv5 model for cow standing recognition is explored[8]; finally, the implemented YOLOv5 model is evaluated and analyzed for environment modeling and target detection algorithm objectives, and the experimental results show that the experimental detection correctness accuracy is 97.6%, and the preprocessing time in detecting a single image is It can quickly and accurately identify the standing behavior of cows, which lays the foundation for basic behavior identification and localization of cows.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125727722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874124
P. Lu
This paper studies the application of wireless transmission of sensor data in coal mine monitoring. Based on LoRa wireless communication technology, wireless sensor network adopts ad hoc network architecture. The low-power design, long-distance transmission and self-organizing network communication of the sensor are studied. Compared with similar technologies, LoRa spread spectrum modulation technology can provide longer communication distance and lower power consumption. LoRa-based wireless sensor network in coal mine adopts clustering tree self-organizing network. The capacity of the network can be dynamically expanded. The network has strong anti-interference ability. This paper expands the application scenario of mobile Internet of Things technology, and improves the intelligent level of coal mine.
{"title":"Design and Implementation of Coal Mine Wireless Sensor Ad Hoc Network Based on LoRa","authors":"P. Lu","doi":"10.1109/ISPDS56360.2022.9874124","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874124","url":null,"abstract":"This paper studies the application of wireless transmission of sensor data in coal mine monitoring. Based on LoRa wireless communication technology, wireless sensor network adopts ad hoc network architecture. The low-power design, long-distance transmission and self-organizing network communication of the sensor are studied. Compared with similar technologies, LoRa spread spectrum modulation technology can provide longer communication distance and lower power consumption. LoRa-based wireless sensor network in coal mine adopts clustering tree self-organizing network. The capacity of the network can be dynamically expanded. The network has strong anti-interference ability. This paper expands the application scenario of mobile Internet of Things technology, and improves the intelligent level of coal mine.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115787471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874022
Wenwu Liu, Chaoqun Zhang, Yunheng Yi, Weidong Qin
With the decrease of farmers and the urgent needs of agricultural modernization, deep learning becomes a novel and effective way to identify crop diseases in modern agriculture. For the problems about low accuracy and complexity of models, a light-weight disease recognition model based on AlexNet is proposed, which is called IAlexNet. The large convolution kernel is replaced by several small convolution kernels to reduce the network parameters, and the SE-Net is introduced to increase the weight of effective information. Besides, the dataset uses the pathological image datasets of apple leaves published on AI studio of the Paddlepaddle. The experiment results show that the recognition accuracy is 97.16%, which is 1.95% higher than AlexNet model. In addition, the parameters of IAlexNet model are reduced by 59.11%, and the training time is reduced by 20.33%, which is verify the new proposed model is feasible and effective.
{"title":"A method for identifying crop diseases based on IAlexNet model","authors":"Wenwu Liu, Chaoqun Zhang, Yunheng Yi, Weidong Qin","doi":"10.1109/ISPDS56360.2022.9874022","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874022","url":null,"abstract":"With the decrease of farmers and the urgent needs of agricultural modernization, deep learning becomes a novel and effective way to identify crop diseases in modern agriculture. For the problems about low accuracy and complexity of models, a light-weight disease recognition model based on AlexNet is proposed, which is called IAlexNet. The large convolution kernel is replaced by several small convolution kernels to reduce the network parameters, and the SE-Net is introduced to increase the weight of effective information. Besides, the dataset uses the pathological image datasets of apple leaves published on AI studio of the Paddlepaddle. The experiment results show that the recognition accuracy is 97.16%, which is 1.95% higher than AlexNet model. In addition, the parameters of IAlexNet model are reduced by 59.11%, and the training time is reduced by 20.33%, which is verify the new proposed model is feasible and effective.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128907163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874106
Jia-nuo Xu, Z. Yao, Jian Yang, Haiyang Wang
The rapid development of consumer UAV has brought many conveniences to national production and life, but the increasingly frequent incidents of ‘Black Flying’ unmanned aerial vehicle (UAV) seriously threaten social stability and national security. In order to capture the operator of ‘Black Flying’ UAV, this paper presents a cross-location method of ground remote controller signal source with the angle of arrival (AOA) as prior information. This method designs a cooperative flight strategy based on double-station UAV airborne detection platform, and builds a mathematical model of cross-positioning error based on this strategy, thus improving the traditional cross-positioning algorithm. The verification results show that the proposed algorithm can successfully locate the signal source of the ground remote controller. Compared with the traditional method, it also effectively reduces the influence of systematic error and random error on the location results. When the angle measurement error is 5 degrees, the location accuracy is about 3 meters.
{"title":"Dual Aircraft Cooperative Cross Positioning Method for Ground Remote Control Signal Source","authors":"Jia-nuo Xu, Z. Yao, Jian Yang, Haiyang Wang","doi":"10.1109/ISPDS56360.2022.9874106","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874106","url":null,"abstract":"The rapid development of consumer UAV has brought many conveniences to national production and life, but the increasingly frequent incidents of ‘Black Flying’ unmanned aerial vehicle (UAV) seriously threaten social stability and national security. In order to capture the operator of ‘Black Flying’ UAV, this paper presents a cross-location method of ground remote controller signal source with the angle of arrival (AOA) as prior information. This method designs a cooperative flight strategy based on double-station UAV airborne detection platform, and builds a mathematical model of cross-positioning error based on this strategy, thus improving the traditional cross-positioning algorithm. The verification results show that the proposed algorithm can successfully locate the signal source of the ground remote controller. Compared with the traditional method, it also effectively reduces the influence of systematic error and random error on the location results. When the angle measurement error is 5 degrees, the location accuracy is about 3 meters.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127311417","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874114
Jiao-yang Li, Chuan Yang, Fan Yang, Jie Huang, Wei Wei, Sujuan Zhang, Xun Zuo, Shilong Zhang
Face detection and tracking technology is used in transportation, security, military fields. In view of the traditional face detection and tracking technology is easy to be affected by light, which leads to low detection accuracy, this paper uses Retinaface and Camshift algorithm to face detection, and realizes real time face detection and tracking by P control steering gear in PID control. Through tests in different environments, the detection accuracy of the Retinaface algorithm and the Camshift algorithm is above 99%. The camera is rotated through P to ensure that the face can be captured by the camera, and the camera response time can reach 0.1s.
{"title":"Face Detection and Tracking Based on Neural Network","authors":"Jiao-yang Li, Chuan Yang, Fan Yang, Jie Huang, Wei Wei, Sujuan Zhang, Xun Zuo, Shilong Zhang","doi":"10.1109/ISPDS56360.2022.9874114","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874114","url":null,"abstract":"Face detection and tracking technology is used in transportation, security, military fields. In view of the traditional face detection and tracking technology is easy to be affected by light, which leads to low detection accuracy, this paper uses Retinaface and Camshift algorithm to face detection, and realizes real time face detection and tracking by P control steering gear in PID control. Through tests in different environments, the detection accuracy of the Retinaface algorithm and the Camshift algorithm is above 99%. The camera is rotated through P to ensure that the face can be captured by the camera, and the camera response time can reach 0.1s.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129462551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874178
Nannan Du, Xuechun Liang, Congyou Wang, Lu Jia
In order to solve the problem of low accuracy of long-term water level forecasting, a multi-station joint long-term water level forecasting model combining random forest and Informer was proposed. First, the Pearson correlation coefficient (PCC) between hydrological stations is calculated, and the hydrological station with the highest degree of correlation with the water level of Hongze Lake is found; then, the random forest (RF) is used to re-extract and select the hydrological station index; finally, the RF and Informer are combined. The experimental results show that the proposed model has higher prediction accuracy.
{"title":"Multi-station Joint Long-term Water Level Prediction Model of Hongze Lake Based on RF-Informer","authors":"Nannan Du, Xuechun Liang, Congyou Wang, Lu Jia","doi":"10.1109/ISPDS56360.2022.9874178","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874178","url":null,"abstract":"In order to solve the problem of low accuracy of long-term water level forecasting, a multi-station joint long-term water level forecasting model combining random forest and Informer was proposed. First, the Pearson correlation coefficient (PCC) between hydrological stations is calculated, and the hydrological station with the highest degree of correlation with the water level of Hongze Lake is found; then, the random forest (RF) is used to re-extract and select the hydrological station index; finally, the RF and Informer are combined. The experimental results show that the proposed model has higher prediction accuracy.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123780563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Aiming at the problem of low image stitching accuracy due to the difficulty of extracting strip features from multispectral images of filter array, a strip stitching algorithm for multispectral images of filter arrays is proposed. Firstly, the effective range of each band of the multispectral image is determined, and the homography matrix is calculated using the geographic coordinate information of the image vertices and the vertex coordinates to project the image. Secondly, Scale-invariant feature transform (SIFT) algorithm was used to extract matching points of projected images, and the mean value of the coordinate difference of matching points was calculated as the translation relation between images. Finally, the projection transformation is performed on the single band bands in turn, and the projected images are stitched with inter-image translation to obtain a large area single band image. Theoretical analysis and experimental results show that this method can effectively improve the the stitching accuracy of multispectral images of filter arrays and has a high image stitching speed.
{"title":"Strip stitching algorithm of filter array multispectral image","authors":"Tong Li, Wenbang Sun, Guang Yue, Zi-lv Gu, Di Wu, Xiaokang Zhang","doi":"10.1109/ISPDS56360.2022.9874187","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874187","url":null,"abstract":"Aiming at the problem of low image stitching accuracy due to the difficulty of extracting strip features from multispectral images of filter array, a strip stitching algorithm for multispectral images of filter arrays is proposed. Firstly, the effective range of each band of the multispectral image is determined, and the homography matrix is calculated using the geographic coordinate information of the image vertices and the vertex coordinates to project the image. Secondly, Scale-invariant feature transform (SIFT) algorithm was used to extract matching points of projected images, and the mean value of the coordinate difference of matching points was calculated as the translation relation between images. Finally, the projection transformation is performed on the single band bands in turn, and the projected images are stitched with inter-image translation to obtain a large area single band image. Theoretical analysis and experimental results show that this method can effectively improve the the stitching accuracy of multispectral images of filter arrays and has a high image stitching speed.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133114435","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874218
Long Ling, Jingde Huang, Yumeng Lu
Deep learning is a hot technology developed in the field of artificial intelligence in recent years. It extracts complex content, simulates the hierarchical structure of the human brain, and constantly adjusts the parameters to find the optimal prediction results. This paper introduces the implementation principle and process of deep learning, uses the deep learning method to study the artifact classification and identification, and completes the artifact classification and identification experiment through the training model of various artifacts. The experimental results show that the sufficient training of the samples can have a high identification accuracy, but the identification accuracy needs to be further strengthened in practical application environments.
{"title":"Study on Artifact Classification Identification Based on Deep Learning","authors":"Long Ling, Jingde Huang, Yumeng Lu","doi":"10.1109/ISPDS56360.2022.9874218","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874218","url":null,"abstract":"Deep learning is a hot technology developed in the field of artificial intelligence in recent years. It extracts complex content, simulates the hierarchical structure of the human brain, and constantly adjusts the parameters to find the optimal prediction results. This paper introduces the implementation principle and process of deep learning, uses the deep learning method to study the artifact classification and identification, and completes the artifact classification and identification experiment through the training model of various artifacts. The experimental results show that the sufficient training of the samples can have a high identification accuracy, but the identification accuracy needs to be further strengthened in practical application environments.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"159 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114567595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874118
Feilong Mao, Yiwen Jiao, Hong Ma, Zefu Gao
High-throughput satellites and satellite Internet constellations are two typical broadband satellite systems. The bandwidth and transmission rate of the system have been greatly improved in recent years. The single-satellite capacity of broadband satellites has increased dozens of times, which brings great challenges to ground receiving equipment. We have sorted out the single-satellite capacity of broadband satellites, the development of Internet satellites, and the speed of satellite communications in recent years. We then conduct a comprehensive review of terrestrial reception equipment for different broadband satellites.
{"title":"Research Status of Broadband Satellite and its Ground Receiving Equipment","authors":"Feilong Mao, Yiwen Jiao, Hong Ma, Zefu Gao","doi":"10.1109/ISPDS56360.2022.9874118","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874118","url":null,"abstract":"High-throughput satellites and satellite Internet constellations are two typical broadband satellite systems. The bandwidth and transmission rate of the system have been greatly improved in recent years. The single-satellite capacity of broadband satellites has increased dozens of times, which brings great challenges to ground receiving equipment. We have sorted out the single-satellite capacity of broadband satellites, the development of Internet satellites, and the speed of satellite communications in recent years. We then conduct a comprehensive review of terrestrial reception equipment for different broadband satellites.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122073364","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874208
Yemao Zhang, Wei Jia, Hai Min, Yingke Lei, Yang Zhao, Chunxiao Fan
Few-shot semantic segmentation aims to tackle the problem that segmenting unseen object class using only a few support images with the same object class. At present, most related methods focus on prototype learning or feature similarity. However, these few-shot segmentation methods do not make good use of high-level features to enhance the prediction results. In this paper, we propose a lightweight Similarity-Guided and Multi-layer Fusion Network (SMNet) with two modules including Similarity-Guided Module (SGM) and Multi-Layer Fusion Module (MLFM). Specifically, the SGM utilizes cosine similarities in multiple high-level feature layers to augment the features in middle-level from query and support image, and then augmented features are refined via a residual attention module. In order to enhance the diversity of features, we reformulate the refined features as a spatiotemporal sequence problem. Then, we introduce the MLFM, which combines two ConvLSTMs to obtain fused feature from different scales. Finally, the decoder takes fused features to obtain predicted mask. Experiment results demonstrate that our model can achieve superior or competitive performances in several datasets.
{"title":"Similarity-Guided and Multi-Layer Fusion Network for Few-shot Semantic Segmentation","authors":"Yemao Zhang, Wei Jia, Hai Min, Yingke Lei, Yang Zhao, Chunxiao Fan","doi":"10.1109/ISPDS56360.2022.9874208","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874208","url":null,"abstract":"Few-shot semantic segmentation aims to tackle the problem that segmenting unseen object class using only a few support images with the same object class. At present, most related methods focus on prototype learning or feature similarity. However, these few-shot segmentation methods do not make good use of high-level features to enhance the prediction results. In this paper, we propose a lightweight Similarity-Guided and Multi-layer Fusion Network (SMNet) with two modules including Similarity-Guided Module (SGM) and Multi-Layer Fusion Module (MLFM). Specifically, the SGM utilizes cosine similarities in multiple high-level feature layers to augment the features in middle-level from query and support image, and then augmented features are refined via a residual attention module. In order to enhance the diversity of features, we reformulate the refined features as a spatiotemporal sequence problem. Then, we introduce the MLFM, which combines two ConvLSTMs to obtain fused feature from different scales. Finally, the decoder takes fused features to obtain predicted mask. Experiment results demonstrate that our model can achieve superior or competitive performances in several datasets.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"124 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122750506","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：