Super-resolution GANs for upscaling unplanned urban settlements from remote sensing satellite imagery - the case of Chinese urban village detection

作     者：Alessandro Crivellari Hong Wei Chunzhu Wei Yuhui Shi 

作者机构：Department of Computer Science and EngineeringSouthernUniversity of Science and TechnologyShenzhenPeople’s Republic of China Ministry of Education Ecological Field Station for East Asian Migratory BirdsDepartment of Earth System ScienceInstitute for Global Change StudiesTsinghua UniversityBeijingPeople’s Republic of China School of Geography and PlanningSun Yat-sen UniversityGuangzhouPeople’s Republic of China Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)ZhuhaiPeople’s Republic of China 

出 版 物：《International Journal of Digital Earth》 (国际数字地球学报（英文）)

年 卷 期：2023年第16卷第1期

页      面：2623-2643页

核心收录：

学科分类：0810[工学-信息与通信工程] 08[工学] 0708[理学-地球物理学] 0816[工学-测绘科学与技术] 081002[工学-信号与信息处理] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the Shenzhen Fundamental Research Program(reference number JCYJ20200109141235597) the National Science Foundation of China(reference number 61761136008)(reference number 42001178) the Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)(reference number 311021018) 

主　　题：SR-GAN resolution upscaling informal settlements urban villages cross-sensor analytics 

摘      要：The semantic segmentation of informal urban settlements represents an essential contribution towards renovation strategies and reconstruction *** this context,however,a big challenge remains unsolved when dealing with incomplete data acquisitions from multiple sensing devices,especially when study areas are depicted by images of different *** practice,traditional methodologies are directed to downgrade the higher-resolution data to the lowest-resolution measure,to define an overall homogeneous dataset,which is however ineffective in downstream segmentation activities of such crowded unplanned urban *** this purpose,we hereby tackle the problem in the opposite direction,namely upscaling the lower-resolution data to the highest-resolution measure,contributing to assess the use of cutting-edge super-resolution generative adversarial network(SR-GAN)*** experimental novelty targets the particular case involving the automatic detection of‘urban villages’,sign of the quick transformation of Chinese urban *** aligning image resolutions from two different data sources(Gaofen-2 and Sentinel-2 data),we evaluated the degree of improvement with regard to pixel-based landcover segmentation,achieving,on a 1 m resolution target,classification accuracies up to 83%,67%and 56%for 4x,8x,and 10x resolution upgrades respectively,disclosing the advantages of artificially-upscaled images for segmenting detailed characteristics of informal settlements.