IMPLEMENTATION OF MACHINE LEARNING APPROACH ON AUTOMATIC RECOGNITION OF HUMAN GAIT PATTERNS

The Biometric authentication is the one that is used to identify the person based on their behavioral and physical activities and characteristics, since it is one of the challenging task that fake biometric products are produced. There are many different techniques of authentication such us face, iris, fingerprint, voice recognition are vulnerabilities of spoofing attacks. Among these human gait recognition is one of the recent research in biometrics field. Gait patterns are used to examine the individual walks, saunters, swaggers with accuracy of about 90 percent. Here in this paper, the captured image is pre-processed and features are extracted from the given image using PCA and LDA algorithm. With the help of the derived features we can classify whether the authentication is correct or not. The Artificial neural network algorithm is used for classification to authenticate the correct person, where the task is completed with the given amount of time so it is found to be of time efficiency.