Automatic shoe sole pattern retrieval system based on image content of shoeprint

The development of a system for automatic retrieval of scene-of-crime shoeprint images from a reference database of shoeprint images is presented in this paper. The approach is based on the image content of shoeprint and a few recent techniques in this area: (1) image probability method for dot texture identification; (2) number of edge's cross dot detect on wavelet edge detect for interlaced texture identification; (3) inflexion's number detect on shape analysis for waving line texture and line texture identification and cross dot's number detect for nubbly texture identification; (4) Hough transform for round texture identification. Experimental results from the application of sole pattern image database are given to show its ability of identifying images and retrieving similar images. The retrieval speed and performance of the system are satisfying.