Vision Based Traffic Personnel Hand Gesture Recognition Using Tree Based Classifiers

Abstract

Human hand gestures can be used as an important communication tool for human computer interaction. As a scientific discipline, computer vision is concerned with the theory behind artificial systems that extract information from images. This paper presents a novel and efficient framework for traffic personnel gesture recognition based on Cumulative Block Intensity Vector (CBIV) of n-frame cumulative difference. The experiment carried out on the real time traffic personnel action dataset using Random Forests (RF) and Decision Tree (J48). Experimental results denote the higher performance 97.83 % of the Random Forests classification, compared to the Decision Tree using 5-frame cumulative difference. The main contribution of this paper is the application of incremental tree based classifier techniques to the problem of identification of traffic personnel hand signals in video surveillance, based only on person hand movement.