Multiple human activity recognition using iot sensors and machine learning in device-free environment: Feature extraction, classification, and challenges: A comprehensive review

Abstract

Human activity recognitions are to identify the actions and goals of one or more individual humans by the internet of things tools from a sequence of observations on the individual human's actions and the environmental conditions. Detection of human activity is a set of techniques that can be used in a wide range of applications, including smart cities, medical health care, and smart home. With the advanced development and commercialization of IoT-enabled devices and crucial demands, human activity measuring the efficient states or health of individuals in smart home-based environments has been a highly dynamic and important topic in recent years since its related to human life. Multi-person activity recognition is a pivotal branch as well as a challenging topic of human action recognition research. However, there are a few methods available for multiple human action recognition based on IoT sensors environment. The objectives of this article are to exploit the multiple human activity recognition systems, overview, feature extraction, classification, challenges, and surveys the state of their art in human activity environment which is based on the IoT sensor and device free as well according to the methodology, datasets, features, limitations used for identifying human behaviors. Finally, the article concludes with an evaluation of the existing approaches, let to frame research problems and questions for future direction which, when applied to real-world scenarios.