An Efficient Classification of Gait Analysis Model using Modified Hybrid Neural Network

Abstract

The Recognition of Gait Identifying people by the way they walk is one of the most under-utilized but effective forms of biometric identification. The premise of this identification method is that each individual has a distinct walk. In addition, it has been widely observed that a person's stride may be used to identify them from a distance if they are familiar with them. Researchers have begun to utilize gait recognition skills due to the growing importance of biometrics in modern personal recognition demands. The purpose of this study is to develop a novel approach to gait detection that use a combination of Artificial Neural Network and Support Vector Machine in order to better understand human gaits (ANNSVM). Background subtraction may be performed in two ways: the first is a recursive technique that uses a Gaussian mixture approach. The second technique is the non-recursive technique, and it employs a sliding-window strategy. Gait recognition consists of a training phase and a testing phase. This paper's concluding portion offers appropriate verification of validation results, presented graphically and with precise description.