{"title":"Object Detection on Aerial Image by Using High-Resolutuion Network","authors":"Zhiyan Bao, Chen Xing, Xi Liang","doi":"10.1109/ICCSNT50940.2020.9304983","DOIUrl":null,"url":null,"abstract":"To detect trespassing in images captured by drones for water conservancy facilities inspection, this paper proposes a method that adapts Hight-Resolution Net to reserve high resolution features for improving detecting results. To detect trespassing target with small scale, this method parallels low-resolution and high-resolution conventical feature maps to reserve high-resolution features, besides that multi-scale fusions are conducted to enhance feature maps with different resolutions. Compare to Faster R-CNN, proposed method achieves 1.7% higher mAP on small targets.","PeriodicalId":6794,"journal":{"name":"2020 IEEE 8th International Conference on Computer Science and Network Technology (ICCSNT)","volume":"7 1","pages":"111-114"},"PeriodicalIF":0.0000,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 IEEE 8th International Conference on Computer Science and Network Technology (ICCSNT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCSNT50940.2020.9304983","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
To detect trespassing in images captured by drones for water conservancy facilities inspection, this paper proposes a method that adapts Hight-Resolution Net to reserve high resolution features for improving detecting results. To detect trespassing target with small scale, this method parallels low-resolution and high-resolution conventical feature maps to reserve high-resolution features, besides that multi-scale fusions are conducted to enhance feature maps with different resolutions. Compare to Faster R-CNN, proposed method achieves 1.7% higher mAP on small targets.