作者: O. El Badawy , M. S. Kamel
DOI: 10.1007/11559573_69
关键词:
摘要: Concavity trees have been known for quite some time as structural descriptors of 2-D shape; however, they haven't explored further until recently. This paper shows how shapes can be concisely, but reversibly, represented during concavity tree extraction. The representation exact, or approximate to a pre-set degree. is equivalent lossless, lossy compression the image containing shape. details proposed technique and reports near-lossless ratios that are 150% better than JBIG standard on test set binary silhouette images.