Abnormal human activity recognition using SVM based approach

Tele-health care applications have gained much attention in the field of ubiquitous computing. With availability of affordable wearable sensors, it is possible to recognize human activities. Activity detection has many applications such as security and health care applications. In this paper, the focus is on detecting abnormal activities of the individuals by ruling out all possible normal activities. Abnormal activities are unexpected events that occur in random manner. Human activities can be recognized using various approaches. Most widely used approach is multi-class SVM. This paper proposes a novel scheme of representing human activities in form of a state transition table. The transition table helps the classifier in avoiding the states which are unreachable from the current state. By avoiding the unreachable states, computational time for classification is reduced significantly when compared to conventional approaches. It is found from simulation studies that the proposed scheme gives accurate results with less computational complexity.