PROJECT TITLE :

Human Detection By Quadratic Classification On Subspace Of Extended Histogram Of Gradients - 2014

ABSTRACT:

This paper proposes a quadratic classification approach on the subspace of Extended Histogram of Gradients (ExHoG) for human detection. By investigating the restrictions of Histogram of Gradients (HG) and Histogram of Oriented Gradients (HOG), ExHoG is proposed as a replacement feature for human detection. ExHoG alleviates the problem of discrimination between a dark object against a bright background and vice versa inherent in HG. It additionally resolves a difficulty of HOG whereby gradients of opposite directions in the identical cell are mapped into the identical histogram bin. We tend to scale back the dimensionality of ExHoG using Asymmetric Principal Component Analysis (APCA) for improved quadratic classification. APCA additionally addresses the asymmetry issue in coaching sets of human detection where there are abundant fewer human samples than non-human samples. Our proposed approach is tested on 3 established benchmarking knowledge sets - INRIA, Caltech, and Daimler - using a modified Minimum Mahalanobis distance classifier. Results indicate that the proposed approach outperforms current state-of-the-art human detection ways.