期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2010
卷号：XXXVIII Part 4
出版社：Copernicus Publications
摘要：High spatial resolution satellite imagery has become an important source of information for mapping and a great number of relatedapplications. Region based segmentation of high resolution imagery is now considered a more suitable method than traditional per pixelclassification techniques. Region growing is a classical method in image segmentation due to its simplicity and effectiveness in makingusing of spatial information among pixels. On the other hand, the automatic and optimal selection of the seeds of growing has been a keyin the context. In order to take great advantage of human vision’s capability of object recognition, this paper presents a semiautomaticsegmentation scheme by which seed regions provided by human operator grow to their boundary separating the seed object and itsbackground. The algorithm ‘learns’ texture measurement from the seed region and tries to expand the seed region till the grown regionhas maximal difference of texture property with the background while the in-class texture property is still consistent. We used a localbinary pattern based texture measurement and tested the approach with a number of high resolution images to extract residential, forestryand different land coverage. The result shows its potential of practical utilization in analysis of high resolution imagery
关键词：High resolution satellite imagery; semiautomatic segmentation; region growing; texture
