基于孪生神经网络的土地利用现状年度变化检测  

Annual change detection of land use status based on siamese neural network

作　　者：张伟良 刘琦 吴长彬[2,3] 杨光迪[2,3] ZHANG Weiliang;IU Qi;WU Changbin;YANG Guangdi(Nanjing Guotu Information Industry Co.,Ltd.,Nanjing 210000,China;School of Geographical Sciences,Nanjing Normal University,Nanjing 210023,China;Key Laboratory of Virtual Geographic Environment,Ministry of Education,Nanjing Normal University,Nanjing 210023,China)

机构地区：[1]南京国图信息产业有限公司,江苏南京210000 [2]南京师范大学地理科学学院,江苏南京210023 [3]南京师范大学虚拟地理环境教育部重点实验室,江苏南京210023

出　　处：《测绘通报》2021年第3期91-95,104,共6页Bulletin of Surveying and Mapping

基　　金：2020年江苏省研究生科研与实践创新计划(SJCX200429)。

摘　　要：近年来,利用多期高分遥感影像,构建土地利用现状年度变化检测模型,并以此为基础支持土地利用现状年度变化检测的智能化发展,正是当前研究的难点所在。本文探讨了构建土地利用现状年度变化检测模型存在的问题。通过对土地资数据进行预处理,构建基于孪生神经网络的变化检测模型,以及对模型输出结果进行GIS优化获取目标区域的变化图斑,实现针对土地利用现状变化区域的快速提取。试验结果表明,该方法可快速发现不同时期影像中土地利用现状的变化位置,有效提升了土地利用现状年度变化检测的智能化水平,可服务于日常土地利用现状年度变更检查工作。In recent years,multi-phase high-resolution remote sensing images are used to build an annual change detection model of land use status and support the intelligent development of annual change detection of land use status on this basis,which is exactly the difficulty of current research. This paper discusses the construction of land use status quo of annual change detection model. Through the land information and data preprocessing,change detection based on twin neural network model being constructed,and the result of the model output optimizing GIS in to obtain the variation spot target area achieves rapid extraction according to current situation of land use change area. The experimental results show that this method can quickly find the change locations of land use status in the images in different periods,which effectively improve the intelligent level of annual change detection of land use status,and can serve the daily annual change detection of land use status.

关 键 词：土地利用现状年度变化检测 自然资源调查监测 影像变化检测 Siamese网络 高分遥感影像 

分 类 号：P237[天文地球—摄影测量与遥感]