Human Activity Recognition: Approaches, Datasets, Applications, and Challenges

Human Activity Recognition (HAR) has been such a demanding problem that needs to be solved. The main focus area of HAR is in healthcare, providing assisted living, especially to elderly people and physically disabled people. This field is getting so much importance as this technique can avoid long hospitalization by monitoring the patients at their residents only. This is done with the help of some other technologies like the Internet of Things (IoT). HAR is made possible by the use of sensors (wearable and non-wearable), smartphones, and images. The data collected by these sensors are stored in the form of a dataset. Then data pre-processing, feature extraction, and classification algorithms are applied to these datasets to get the desired outcome. In this chapter, different publicly available datasets are discussed along with the approaches that can be used to deal with HAR, then its different applications and the challenges that are being faced in the proper execution of HAR systems have been discussed.