Paper Details

Activity recognition is formulated as a classification problem. Human activity recognition (HAR) receives moreattentions in recent years, due to many of its applications, such as health care, video surveillance and context-awarecomputing. Human activity are recognized from accelerometer data. Now a days, smartphone are used intensively so the keybenefits of using the smartphone accelerometer for human mobility analysis, with or without location determination basedupon GPS, WiFi or GSM is that accelerometer is energy-efficient, provides real-time contextual information and has highavailability. Using measurements from an accelerometer for human mobility analysis presents its own challenges as all carrysmartphones differently and the measurements are body placement dependent. Therefore a novel algorithm is proposed thatneutralizes the effect of different smartphone on-body placements and orientations to allow human movements to be moreaccurately and energy-efficiently identified. We proposed android application AT (Activity Tracker) to collect dataset whichruns on android and iOS, record accelerometer data with 40 Hz frequency like 40 samples per seconds.

Ms. Puja B. Patil, Mrs. D. B. Gothawal, “Real-Time Triaxial Accelerometer-based Energy Efficient Human Activity Recognition using Smartphone”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 4, Issue 6, pp. 618-624, June 2017.