{"title":"基于双向CNN的无人机对地目标鲁棒跟踪算法","authors":"Zhijun Liu, Di Zhang","doi":"10.1109/PHM-Yantai55411.2022.9941861","DOIUrl":null,"url":null,"abstract":"In order to make the UAV equipment track the ground target accurately in the flight process, this paper proposes a UAV-To-Ground target robust tracking algorithm based on Two-Way CNN. The algorithm is based on two-way CNN architecture, and detects the visual target of UAV after solving the surf feature of the ground target image. According to the linear correction expression of the tracking parameters, the algorithm determines the performance intensity of the non maximum suppression effect of the tracking coefficient on UAV ground target parameters, and then combines the known tracking coefficient and the loss function to realize the tracking of the UAV ground target. The experimental results show that under the effect of the dual CNN network architecture, the tracking accuracy of the UAV equipment for the established ground target during flight is significantly improved, which can meet the actual application requirements.","PeriodicalId":315994,"journal":{"name":"2022 Global Reliability and Prognostics and Health Management (PHM-Yantai)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"UAV-to-Ground Target Robust Tracking Algorithm Based on Two-Way CNN\",\"authors\":\"Zhijun Liu, Di Zhang\",\"doi\":\"10.1109/PHM-Yantai55411.2022.9941861\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In order to make the UAV equipment track the ground target accurately in the flight process, this paper proposes a UAV-To-Ground target robust tracking algorithm based on Two-Way CNN. The algorithm is based on two-way CNN architecture, and detects the visual target of UAV after solving the surf feature of the ground target image. According to the linear correction expression of the tracking parameters, the algorithm determines the performance intensity of the non maximum suppression effect of the tracking coefficient on UAV ground target parameters, and then combines the known tracking coefficient and the loss function to realize the tracking of the UAV ground target. The experimental results show that under the effect of the dual CNN network architecture, the tracking accuracy of the UAV equipment for the established ground target during flight is significantly improved, which can meet the actual application requirements.\",\"PeriodicalId\":315994,\"journal\":{\"name\":\"2022 Global Reliability and Prognostics and Health Management (PHM-Yantai)\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-10-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 Global Reliability and Prognostics and Health Management (PHM-Yantai)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/PHM-Yantai55411.2022.9941861\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 Global Reliability and Prognostics and Health Management (PHM-Yantai)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/PHM-Yantai55411.2022.9941861","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
UAV-to-Ground Target Robust Tracking Algorithm Based on Two-Way CNN
In order to make the UAV equipment track the ground target accurately in the flight process, this paper proposes a UAV-To-Ground target robust tracking algorithm based on Two-Way CNN. The algorithm is based on two-way CNN architecture, and detects the visual target of UAV after solving the surf feature of the ground target image. According to the linear correction expression of the tracking parameters, the algorithm determines the performance intensity of the non maximum suppression effect of the tracking coefficient on UAV ground target parameters, and then combines the known tracking coefficient and the loss function to realize the tracking of the UAV ground target. The experimental results show that under the effect of the dual CNN network architecture, the tracking accuracy of the UAV equipment for the established ground target during flight is significantly improved, which can meet the actual application requirements.