A Novel Automated Shoeprint Matching Technique for use as Forensic Evidence in Criminal Investigation

Abstract

This work presents a technique in which a database of outsole pattern of shoeprint images has been automatically sorted against a query shoeprint image. Shoe marks found at the place of crime are used to provide valuable forensic evidence. This system presents a technique for rotation and intensity invariant automatic shoeprint matching so that the spatial positioning of the reference shoeprint image does not have to correspond with the spatial positioning of the shoeprint images of the database. Gabor transform has been used to extract multi resolution features of a shoeprint. Radon transform has been used to estimate the rotation of the shoeprint image and is compensated by rotating the features in opposite direction. Shoeprint database has been generated by inviting participants to tread on an inkpad and then stamp on a piece of paper. Template query images have been compared using Euclidean distance classifier which has been used to find a suitable match. The performance of the proposed algorithm has been evaluated in terms of percentage accuracy for four different matching methods. This technique performs better compared to results obtained using power spectral density features for full print images with the rotation, intensity and mixed attacks. General Terms Image processing, Forensic application.