A facial recognition system for matching computerized composite sketches to facial photos using human visual system algorithms

Forensic sketching techniques are a commonly used method to assist law enforcement agencies. With the powerful advances in computer software, computerized composite sketches are more frequently used than hand-drawn sketches because of its efficiency and low-cost. However, there is not much research focused on facial recognition systems for matching computerized composite sketches to facial photos. This problem is addressed by introducing an automatic recognition system for matching computerized facial sketches to face photos utilizing human visual system algorithms, coupled with logarithm Logical Binary Pattern feature descriptors and a new region weighted model. The architecture can quickly find and rank the closest matches of a test image to a database of stored images. The contributions of this paper include: (i) A database containing people of interest with corresponding forensic images and their computerized sketches that will be available for future researchers; (ii) An automatic computerized sketch to facial photo recognition system utilizing human visual system algorithms; (iii) This work will also compare the accuracy between matching forensic photos and matching the composite sketches. The results show the high efficiency and accuracy of the prototype system. The proposed computerized sketch matching facial photo system has potential for many law enforcement and homeland security applications.