Semantics-sensitive Retrieval for Digital Picture Libraries
James Ze Wang, Jia Ming Li, Desmond W. Chan, Gio C. M. Wiederhold · D-Lib Magazine · 1999
This paper presents an algorithm to segment images into four classes: background, photograph, text and graph.There are two important aspects about the algorithm.The rst is that the algorithm takes a multiscale approach, which adaptively classi es an image at dierent resolutions.The multiscale structure enables accurate classi cation at class boundaries as well as fast classi cation overall.The second is that the context information, which is accumulated in the process of classi cation, is used to improve the classication accuracy.