Gait based gender classification using neural and non-neural

Abstract: Gait as one of the behavioural biometric recognition aims to recognize an individual by the way they walk. In this paper we propose gender classification based on human gait features based on wavelet transform and investigates the problem of non-neutral gait sequences: Coat Wearing (CW) and carrying bag(CB) condition as addition to the neutral gait sequences. The proposed method is based on a new set of feature constructed based on the Gait Energy Image and Gait Entropy Image called Gait Entropy Energy Image (GEnEI). Three different feature sets constructed from GEnEI based on wavelet transform called, Approximation coefficient Gait Entropy Energy Image (AGEnEI), Vertical coefficient Gait Entropy Energy Image (VGEnEI) and Approximation & Vertical coefficients Gait Entropy Energy Image (AVGEnEI). Finally two different classification methods are used to test the performance of the proposed method separately, called k-nearest-neighbour (k-NN) and Support Vector Machine (SVM). Our tests are based on large number of experiments using CASIA B gait database, includes 124 subjects (31 women and 93 men). The experimental result shows that the results achieved by the method outperform the state of the art.