Human activity recognition in real world

Human activity recognition is an important area of computer vision research. The goal of activity recognition is an automated analysis or interpretation of ongoing events and their context from video data. Application of activity recognition is used for surveillance system, patient monitoring system, and a variety of system that involve interaction between person and electronic devices such as human computer interfaces. The prime objective of this research paper is to recognize the human activity in the real world. To improve the performance of activity recognition system we will apply K-means clustering including pruning. Frames is extracted from videos in the form of image using function frame2im. For better recognition further features is extracted from these frames using LDA. Then these extracted frames are used to train in neural network which in results shows the improvement in recognition system.