Target identification for photon-counting image sensors, inspired by mechanisms of human visual perception

Abstract

Abstract Paper presents the new objects identification/recognition method, designed for modern image sensors. One of the main features of the method proposed is that it is initially focused on presenting the input data as a discrete set of photocounts of image sensors. The developing of the method was based on the machine learning approaches, so the synthesized identification algorithms are principally focused on the modern embedded computations. The results of numerical simulation are given at the end of the paper.