Guoqi Liu, Manqi Zhao, Lu Bai, Hecang Zang, Baofang Chang
{"title":"Field road segmentation network based on PraNet","authors":"Guoqi Liu, Manqi Zhao, Lu Bai, Hecang Zang, Baofang Chang","doi":"10.1080/14498596.2022.2059023","DOIUrl":null,"url":null,"abstract":"ABSTRACT Nowdays, many methods based on CNN have been proposed for road extraction. However, there are still great challenges. Therefore, according to image characteristics, this paper made corresponding improvements based on medical segmentation network PraNet. First, the reverse attention module (RA) connected at the last layer of PraNet is changed to the positive attention module (PA). Then, the negative matrix L1 norm regularization is added into the loss function. We conducted experiments on a data set made by UAV in the field of Henan Academy of Agricultural Sciences. The results show that the proposed method is better than the comparison models.","PeriodicalId":50045,"journal":{"name":"Journal of Spatial Science","volume":"68 1","pages":"505 - 520"},"PeriodicalIF":1.0000,"publicationDate":"2022-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Spatial Science","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1080/14498596.2022.2059023","RegionNum":4,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"GEOGRAPHY, PHYSICAL","Score":null,"Total":0}
引用次数: 0
Abstract
ABSTRACT Nowdays, many methods based on CNN have been proposed for road extraction. However, there are still great challenges. Therefore, according to image characteristics, this paper made corresponding improvements based on medical segmentation network PraNet. First, the reverse attention module (RA) connected at the last layer of PraNet is changed to the positive attention module (PA). Then, the negative matrix L1 norm regularization is added into the loss function. We conducted experiments on a data set made by UAV in the field of Henan Academy of Agricultural Sciences. The results show that the proposed method is better than the comparison models.
期刊介绍：
The Journal of Spatial Science publishes papers broadly across the spatial sciences including such areas as cartography, geodesy, geographic information science, hydrography, digital image analysis and photogrammetry, remote sensing, surveying and related areas. Two types of papers are published by he journal: Research Papers and Professional Papers.
Research Papers (including reviews) are peer-reviewed and must meet a minimum standard of making a contribution to the knowledge base of an area of the spatial sciences. This can be achieved through the empirical or theoretical contribution to knowledge that produces significant new outcomes.
It is anticipated that Professional Papers will be written by industry practitioners. Professional Papers describe innovative aspects of professional practise and applications that advance the development of the spatial industry.
京公网安备 11010802042870号
京ICP备2023020795号-1
分享
求助内容：
应助结果提醒方式：
扫码关注我们
