Implementation of Human Walking Action GAIT Recognition Using Hidden Markov Model and Radial Basis Function Neural Network

This paper presents the combined implementation of radial basis function(RBF) along with hidden Markov model (HMM) for human activity recognition. Surveillance cameras are installed in the crowded area in major metropolitan cities in various countries. Sophisticated algorithms are required to identify human walking style to monitor any unwanted behavior that would lead to suspicion. This paper presents the importance of RBF to identify the human GAIT. GAIT is one of the biometrics that can be measured at a distance and useful for security surveillance and biometric applications. The attraction of using GAIT as a biometric is that it is non-intrusive and typifies the motion characteristics specific to an individual. The proposed system attempt to recognize people by modeling each individual's GAIT using HMM. The HMM is a good choice for modeling a walk cycle because it can model sequential processes. This knowledge is used to generate a lower dimensional observation vector sequence which is then used to design a continuous density HMM for each individual