Face Sketch-Image Recognition for Criminal Detection Using a GAN Architecture

One of the important cues in solving crimes and apprehending criminals is matching sketches with digital face images. The problem of matching a forensic sketch to a gallery of mugshot images is addressed in this paper. Feature-based technique is implemented and compared against proposed generative adversarial networks (GANs). The designed GAN projects alternate accuracy for the generator and the discriminator for various batch sizes, dropouts, and learning rate and is able to identify the corresponding image in the CUHK database. Simulation results show that while the feature-based matching fails the holistic method having been trained as an innovation process produced promising results.