高分辨率遥感影像道路分割算法研究-摄影测量与遥感专业论文.docxVIP

AbstractThe印plication Abstract The印plication of high resolution remotely sensed data has come into new era since SPOT．5，IKONOS and QuickBird sateIlites were launched．As a result，new methods are required to ob诅in infomlation丘om the high—resolution imagery． Extracting road features automatically丘Dm remote sensing images haS pmVed to be a di箍cuIt task．n is aIso one of the key problems of photogrammetry and化mote sensing，which plays a cmcial role on photogrammetry aIId automating the analysis of the remote sensing image．A binary road image can be obtained by means of segmentation，aIld eXtracting mads in such an image is ofcourse easier t11an itl a gray one．In fact，extracting roads in a binary image is a represematiVe hnd of methDds In mis papefwe mainly pay our attention to me segmentation algoritllIIls ofthe road images．The印ecinc characteristics of me typical urban mads in t11e mgh fesolution remote sensing images for typical urban roads are considered，ajld me corresponding algorithms have been used to distinguish the road打om other objects This pmcess provides the preparation for the丘11al extraction．Major research contems are tlle following： (1)Some of the existing segmentation appmaches for hi曲resolution remote sensing images have been analyzed in detail，and their adVantages and disadVantages have been summed up． (2)The segmentation algorimm for multi—category target images which is based on 0TSU has been researched carefully，and its lim“ations while being used to segment roads directly has been summarized as well． (3)Two new methods suitable for t}le urban high resolution remote sensing images have been proposed．The first is base on me gray consistency of the regional road and the second is based on the inte掣ated features． The method based on me gray consistency of me regionaI road uses a combination of gray and space position infomation to segment images． Experimental results show that tms method is more e爿’ectiVe，as it caI】be used to deal with the complex hi曲resolut