Human Gait Recognition based on Principal Curve Component Analysis

As a biometric technology, gait has recently gained more and more interests from computer vision researchers. A gait recognition algorithm based on principal curve component analysis was proposed. Principal curve component analysis can model nonlinear data effectively, which analyzes the data from its inherence and emphasizes the nonparametric characteristic. First, a background subtraction was used to separate objects from background. Then, we represented the silhouette features based on moments and extracted the variety of gait sequences. Finally, we used principal curve component analysis to analyze gait features. The performance of our approach was tested using different gait databases. Our approach shows a better recognition rate. Principal curve component analysis can analyze nonlinear gait data effectively