Performance Analysis of Face Matching and Retrieval In Forensic Applications

This paper highlight the challenges in applying face recognition technology in forensic applications. It addresses two specific research problems 1) face retrieval using various pose and expressions 2) matching forensic sketches to face photograph databases. Solutions to these problems are necessary to accurately remove the duplicates in various government databases, including passport, driver license photos; to apprehend the criminals when no photos of suspect is available (sketch to photo matching).This can be done by using several algorithms. For pose and expression, Wavelet Transform, Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) algorithms are used.