A hierarchical abnormal human activity recognition system based on R-transform and kernel discriminant analysis for elderly health care

A hierarchical human activity recognition (HAR) system is proposed to recognize abnormal activities from the daily life activities of elderly people living alone. The system is structured to have two-levels of feature extraction and activity recognition. The first level consists of R-transform, kernel discriminant analysis (KDA), k-means algorithm and HMM to recognize the video activity. The second level consists of KDA, k-means algorithm and HMM, and is selectively applied to the recognized activities from the first level when it belongs to the specified group. The proposed hierarchical approach is useful in increasing the recognition rate for the highly similar activities. System performance is analyzed by selecting the optimized number of features, number of HMM states and the number of frames per second to achieve maximum recognition rate. The system is validated by a novel set of six abnormal activities; falling backward, falling forward, chest pain, headache, vomiting, and fainting and a normal activity walking. Experimental results show an average recognition rate of 97.1 % for all the activities by using the proposed hierarchical HAR system.