Assessment of Image Enhancement Procedures for Matching Sketches to Photos

Abstract

Image supported examination procedures are widely considered in the variety of fields. Typically, the image examination involves in the extracting the region of interest or features for the further investigation. In human detection system and also in criminal science, identification of the correct person is essential for the authentication requirements. In these applications, usually, the sketch drawn by an expert or a computer is matched against the digital photographs available in the criminal or the public database. During this examination, essential facial features are extracted from the sketch and digital photography is compared to identify the percentage of fit. The proposed work in this paper aims to examine the existing image enhancement and identification procedures existing in the literature. After creating the essential sketch for a person, a comparative analysis against the digital photo (Gray / RGB scale) of the person is performed, and the image similarity measures, such as PSNR, MSE, and SSIM are computed. In this paper, the benchmark datasets, like CUHK and PRIP-HDC are considered for the examination, and the empirical trial is achieved with the help of Matlab software.