Threshold Value Determination for Recognition of Partial Shoe Prints for Forensic Analysis

This paper aims to develop an automated system to aid forensic scientists in rapid identification of the partial shoe print images based on Phase only Correlation technique. The proposed new technique captures more discriminative information when compared to amplitude based methods. The main advantage of this technique is the capability to match a low quality shoe print image accurately and efficiently. The Cross-Phase Spectrum function coupled with the Spectral Weighting function results in high performance and increased recognition rate method. The proposed algorithm was checked on real time images obtained on various test beds. To check the efficiency of the algorithm the percentage of matching between the reference image and the sample image were performed.