<![CDATA[Identifying a person based on his or her characteristics is the work done on the scope of biometrics. For example each individual has a unique characteristic such as a face or fingerprint. Face is the most recognizable feature which used in the field of forensics only by knowing the physical features on its face by witnesess. Discriminative approach used in this experiment is a well-known method, SIFT (Scale Invariant Feature Transform) and MCWLD (Multiscale Circular Weber Local Descriptor)/ Starting from extracting local features from a set of photos and sketches then performed matching using euclidean distance. The result of this experiment proves that SIFT method with configuration of small window size 8 and 32-overlapping sliding window achieves 79.79% identification rate in top-match rank, while for MCWLD with 16-overlapping sliding window configuration and the parameters used are T = 6, M = 4 and S = 3 achieved 82.45% identification rate on top-match rank. Although MCWLD's identification rate are better than SIFT on top-match, but on the overall result or top-rank, SIFT's identification over-perform MCWLD.]]>
Copyrights © 2018
