Shape Retrieval with Flat Contour Segments

摘要: image database, retrieval, run length coding, Freeman downsampling Content-based retrieval (CBIR) is an important issue in the computer vision community. Both visual and textual content descriptions are employed when user formulates queries. Shape feature feature, as it corresponds to region of interest images. For shape comparisons must be compact accurate, invariant several geometric transformations such translation, rotation scaling, even if particular representation may rotated. In this paper, we propose a comparison technique based on flat segments contour. The segmentation utilizes coding coding. lengths make up vector, which used compare similarity shapes. Experimental results from test standard SQUID database reported.