LAUSR

In recent years, the recognition of activity is a daring task which helps elderly people, disabled patients and so on. The aim of this paper is to design a system for recognizing the human activity in egocentric video. In this research work, the various textural features like gray level co-occurrence matrix and local binary pattern and point feature speeded up robust features are retrieved from activity videos which is a proposed work and classifiers like probabilistic neural network, support vector machine (SVM), k nearest neighbor (kNN) and proposed SVM+kNN classifiers are used to classify the activity. Here, multimodal egocentric activity dataset is chosen as input. The performance results showed that the SVM+kNN classifier outperformed other classifiers.