跨视角图像地理定位方法综述  

Review of cross-view image geolocalization methods

作　　者：盛怡宁 赵理君 张正 崔绍龙 饶梦彬 唐娉 Sheng Yining;Zhao Lijun;Zhang Zheng;Cui Shaolong;Rao Mengbin;Tang Ping(Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China;School of Electronic,Electrical and Communication Engineering,University of Chinese Academy of Sciences,Beijing 100049,China;Information Science Academy of China Electronics Technology Group Corporation,Beijing 100043,China)

机构地区：[1]中国科学院空天信息创新研究院,北京100094 [2]中国科学院大学电子电气与通信工程学院,北京100049 [3]中国电子科技集团公司信息科学研究院,北京100043

出　　处：《中国图象图形学报》2024年第9期2716-2736,共21页Journal of Image and Graphics

基　　金：中国科学院空天信息创新研究院“未来之星”人才计划项目(2020KTYWLZX03,2021KTYWLZX07);中国科学院青年创新促进会项目(2022127)。

摘　　要：跨视角图像地理定位旨在通过图像匹配和地理坐标估计实现不同视角图像之间的准确对应和地理定位,广泛应用于机器人导航、自动驾驶和三维重建等领域。传统的单一视角图像地理定位方法通常受限于数据集质量和规模等因素,定位精度较低。为克服这些局限,近年来研究人员提出了一系列跨视角图像地理定位方法,同时利用多个视角的图像数据,通过视角比较和匹配提高定位精度。跨视角图像匹配方法呈现多元的分类体系。根据面向的跨视角图像类型的不同,可将其分为面向地面—卫星图像的方法与面向无人机—卫星图像的方法两类。根据图像特征提取与表达方式的不同,又可将其分为基于人工设计特征的方法与基于深度神经网络自学习特征的方法两类,对于后者,还可根据是否采用视角对齐方法以及所采用对齐方法的不同将其细分为无视角对齐处理的跨视角图像地理定位、基于传统图像变换的跨视角图像地理定位和基于图像生成的跨视角图像地理定位等3类。本综述对以上方法进行了介绍并比较了它们的优缺点;此外,还总结了常用于跨视角图像地理定位的数据集和评价方法;最后,展望了跨视角图像地理定位的应用领域和未来发展方向。尽管跨视角地理定位方法已取得突破和进展,但仍面临一些问题和挑战。因此,本综述提出了可能的解决方向和未来研究的重点,以期推动该领域的发展和创新。The research field of cross-view image geolocalization aims to determine the geographic location of images obtained from various viewpoints or perspectives to provide technical support for subsequent tasks,such as automatic driv⁃ing,robot navigation,and three-dimensional reconstruction.This field involves matching images captured from different views,such as satellite and ground-level images,to accurately estimate their geographical coordinates.Cross-view image geolocalization presents difficulty due to differences in viewpoint,scale,illumination,and appearance among images.This process requires addressing the problems of viewpoint variation,geometric transformations,and handling the large search space of possible matching locations.Early studies on image geolocalization were mainly based on single-view images.Single-view image geolocalization can obtain the geolocation information of a given image by searching for the sameview reference image with prelabeled geolocation information from the image database.However,the traditional singleview image geolocalization method is usually limited by the quality and scale of the dataset,and thus,the positioning accu⁃racy is usually low.To overcome these limitations,the researchers have proposed a series of cross-view image geolocaliza⁃tion methods that utilize image data from multiple perspectives to increase the positioning accuracy through the comparison and matching various perspectives.Given the complexity of geolocalization tasks and solutions,existing methods of crossview image geolocalization can be classified in multiple ways.This review introduces various classification methods of cross-view image geolocalization and representative methods for each type,and compares their advantages and disadvan⁃tages.On the one hand,the diversification of platforms and the increase in multisource data provide more source data choices for cross-view image geolocalization.Based on the differences in matching image sources,cross-view image geolo⁃calization methods can be c

关 键 词：图像地理定位 跨视角 图像匹配 深度学习 表征学习 视角转换 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]