基于Geoway Feature的基础测绘变化信息检测研究  

Study on Basic Surveying and Mapping Change Information Detection Based on Geoway Feature

作　　者：刘一锋 LIU Yifeng(Mapping Institute of Gansu Province,Lanzhou 730000,China)

机构地区：[1]甘肃省地图院,甘肃兰州730000

出　　处：《地理空间信息》2023年第8期126-128,共3页Geospatial Information

基　　金：甘肃省自然科学基金资助项目(20JR10RA452)。

摘　　要：随着各领域对基础地理信息现势性需求的增强,基础测绘工作需实现变化要素的实时处理,因此变化信息检测的效率与精度成为制约基础测绘更新效率的瓶颈。以提高变化信息提取的自动化程度、促进靶向性更新的快速开展为目标,依托Geoway Feature遥感信息提取软件,采用基于深度学习策略的计算机变化信息检测方法,对甘肃省庆城市区域多时相卫星正射影像展开试验,最后对实验效能予以评价,对其应用前景、研究方向进行了展望。结果表明Geoway Feature软件的检测结果的准确性与检测对象种类、样本库数量质量相关性大。With the increasing requirement of basic geographical information in various fields,basic surveying and mapping needs to achieve real-time processing of change elements.Therefore,the efficiency and accuracy of change information detection become the bottleneck for the update efficiency of basic surveying and mapping.In order to improve the automation of change information detection and promote the rapid development of targeted update,basedonGeowayFeature,using the computer change discovery method based on deep learning strategy,we carried out experiment on multi-temporal satellite orthographic images in Qingcheng City,Gansu Province.Finally,we evaluated the effectiveness of this experiment,and prospected its application and research direction.The experimental resultshows the detecting accuracy of GeowayFeature is highly related to the type of detected object,the quantity and quality of sample library.

关 键 词：深度学习 变化信息检测 基础测绘 Geoway Feature 

分 类 号：P25[天文地球—测绘科学与技术]