Review on Real Time Approach to Identify a Person based on Hand drawn Sketch Using Deep Learning

To identify a criminal in law enforcement, facial images collected by forensic sketch artists are used. These sketches play a crucial role in identifying a criminal or missing individual if no pre-existing database is accessible. The existing recognition systems are based on a mugshot database, which comprises images of criminals. This research work presents a novel method based on a real-time technique for identifying a person from a live video using free hand-drawn sketches created using eyewitness information. Deep learning for forensic sketch recognition is an important area of research in computer vision. The deep learning, Convolution Neural Network (CNN) algorithm is used to identify human faces from a live video and extract various features from an actual image and sketch. Siamese neural network is a deep learning model that is trained to compare two images and learn about the similarity metrics between two images to select it. This real-time approach leverages enhanced performance.