Sensors-Based Ambient Assistant Living via E-Monitoring Technology

Independent human living systems require smart, intelligent, and sustainable online monitoring so that an individual can be assisted timely. Apart from ambient living, the task of activities plays important role in different fields including virtual reality, surveillance security, interaction with robots. Such have been developed past use various wearable inertial sensors depth cameras to capture actions. In this paper, we propose multiple methods such as random occupancy pattern, spatio temporal cloud, way-point trajectory, Hilbert transform, Walsh Hadamard transform bone pair descriptors extract optimal features corresponding These sets are then normalized using min-max normalization optimized Fuzzy optimization method. Finally, Masi entropy classifier is applied for action recognition classification. Experiments performed on three challenging datasets, namely, UTD-MHAD, 50 Salad, CMU-MMAC. During experimental evaluation, proposed novel approach recognizing actions has achieved accuracy rate 90.1% UTD-MHAD dataset, 90.6% Salad 89.5% CMU-MMAC dataset. Hence results validated system.