Image processing and artificial neural network for counting people inside public transport

Abstract

In this paper, we present a system for counting people on public transport. The system consists of a single camera, and it process videos obtained from the camcorder to perform an algorithm to identify and to count people using Artificial Neural Networks (ANNs). The system is implemented by Cpp language based on OpenCV. People counting is an important factor to guarantee safe road trips as transport vehicles must not exceed an established number of passengers. This research is aimed to create a technological tool that helps to comply with the Transit Law in Ecuador. We describe different processing blocks dedicated to the detection, segmentation, direction estimation, counting, and evaluation of images to present its results during the standard travel of a bus.