Experimental Assessment on Latent Fingerprint Matching Using Affine Transformation

In forensics latent fingerprint identification is critical importance to identifying suspects: latent fingerprints are invisible fingerprint impressions left by fingers on surfaces of objects. The proposed algorithm uses a robust alignment algorithm (mixture contour and Orientation based Descriptor) to align fingerprints and to get the similarity score between fingerprints by considering minutiae points and ridge orientation field information.The texture-based descriptors (local binary patterns and local phase quantization), address important issues related to the dissimilarity representation, such as the impact of the number of references used for verification and identification. However, the overlapped region shape similarity retrieved from minutiae spatial distribution information provides additional important criteria. After finding the overlapping region of a possible affine transform, we can measure to find the shape dissimilarity via the application of the shape context to all interior points.TheHybrid matching algorithm, is to prune outlier minutiae pairs, and secondly to provide more information to use in similarity evaluation. Keywords—Orientation and texture descriptors, latent‘s, affine transform, contour context, matching.