Human Gait Recognition Using Multisvm Classifier

The human gait is an important biometric feature for human identification. Individuals have distinctive and special ways of walking. This methodology comes under the model free approaches, different types of features (e. g. , the whole silhouettes silhouette width, height values and joint angles) are first extracted. In the subsequent pattern-matching stage, some approaches exploit the silhouette shape and dynamics information. By the usage of model free approach, it can reduce the dimensionality which in turn reduces time complexity. Nowadays human gait recognition has an increasing research interest in human identification in controlled environments such as airports, banks, and car parks. The human gait is an important feature for human identification in such video surveillance-based applications because it can be perceived unobtrusively from a medium to a great distance. This system uses a new patch distribution feature (PDF) for human gait recognition. It represent each gait energy image (GEI) as a set of local augmented Gabor features, which concatenate the Gabor features extracted from different scales and different orientations together with the X-Y coordinates. Then learn a global Gaussian mixture model (GMM) with the local augmented Gabor features from all the gallery GEIs; then, each gallery or probe GEI is further expressed as the normalized parameters of an image-specific GMM adapted from the global GMM. To enhance the accuracy and to reduce the time complexity of this system a new classification method using multisvm classifier is also combined to this method.