A system for counting people in video images using neural networks to identify the background scene

Abstract A method for counting the number of people in any pre-defined scene is described. The method has three distinct stages: image pre-processing, background identification and object search. The method was designed to provide accurate counts, even when the background scene was allowed to vary. This tolerance to changes in the background scene was achieved using RAM-based neural network classifiers to identify sections of the background scene in each test image. The system was implemented on relatively low cost hardware and was found to give good results at moderately high frame rates.