Abstract
Recently, considerable measures are taken to improve security in various locations. Most of the methods use the image-supported identification due to its accuracy and robustness. Usually, the picture assessment engages in capturing the raw image, identifying the vital information, extracting the features, and classification and authentication. This technique is used widely in human recognition and also in criminal science, recognition of the right individual is necessary for the verification necessities. In these applications, usually, the sketch drawn by an expert or a computer is matched alongside the digital photographs accessible in the criminal or the public database. Throughout this assessment, necessary facial features are mined from the photo and compared it with the original picture. This paper aims to evaluate the existing picture improvement and recognition procedures in the literature. After generating the necessary drawing for an individual, a relative examination against the digital picture of the person is executed and the image similarity measures are computed to authenticate photo with the sketch. This work proposes a methodology to evacuate the proposed technique using benchmark datasets and the result demonstrates a better result.
