Pedestrian detection using integral channel detection and ADABOOST algorithm
Kavita Wagh, Sudhir S. Kanade · 2017
Pedestrian detection and tracking is a challenging task in surveillance video system as there are variations in appearance, poses, color, shadow, interference. Past study reveals that several schemes have been developed for pedestrian detection and tracking, however there is a need for improvement in accuracy and robustness. This paper proposes a new concept to improve these measures to remove variations, in which the essential process is to extract features color, HOG and gradient as Integral Channels Features. These features form the estimated model from the feature and the estimated model is used as a detector for pedestrian detection using ADABOOST algorithm. From the results shown in this paper, it is clear that the technique of integral channel features and ADABOOST algorithm used for detection combine gives very prominent results for pedestrian detection. An experimental result illustrates the improvement in efficiency and is highly competitive in terms of misclassification error rate.