A customized 1D-CNN approach for sensor-based human activity recognition
International Journal of Advanced Technology and Engineering Exploration · 2022
Human activity recognition (HAR) is an emerging area of research that is applicable in several fields such as health care, security, smart homes, fitness assessment, sports competition and assisted living.HAR is performed by acquiring data from wearable sensors and external sensors like cameras, pressure sensors, and video recorders.Vision-based HAR plays an active role in the detection of suspicious activity and events in video surveillance.The requirement of computationally expensive image and video analytics, expensive hardware, and lack of portability are the main limitations of vision-based HAR [1].