Human Gait Recognition Based on Kernel PCA Using Projections

This paper presents a novel approach for human identification at adistance using gait recognition. Recognition of a person from theirgait is a biometric of increasing interest. The proposed work introducesa nonlinear machine learning method, kernel PrincipalComponent Analysis (PCA), to extract gait features from silhouettes forindividual recognition. Binarized silhouette of amotion object is first represented by four 1-D signals which are thebasic image features called the distance vectors.Fourier transform is performed to achieve translationinvariant for the gait patterns accumulated from silhouette sequenceswhich are extracted from different circumstances.Kernel PCA is then used to extracthigher order relations among the gait patterns for future recognition. Afusion strategy is finally executed to produce a final decision.The experiments are carried out on the CMU and the USF gait databasesand presented based on the different training gait cycles.