In the past decade, human activity recognition (HAR) has evolved into a very active research topic with its potential applications in many fields, such as surveillance, smart environments, human-machine interaction, image and video retrieval, and healthcare. For instance, it can be adopted in a healthcare service system to monitor the rehabilitation processes of patients and provide useful information for clinicians to help them make decisions for clinical diagnoses, patient monitoring, and treatment management.
Recently, three trends are fueling the development of HAR. First of all, ubiquitous (non-invasive) sensors provided by the Internet of things (IoT) and smart phones, etc., generate an ever-increasing amount of data for the recognition of human activities. Secondly, new classes of algorithms, in particular in the field of machine learning, such as deep learning convolution networks and Neural Networks, enable a better automated understanding of the data. Third, computation power not available in previous decades is now a reality even on edge devices. Therefore, entirely new applications of HAR are becoming possible due to the ubiquity of sensors and the availability of new classes of algorithms.
This special issue intends to prompt emerging techniques on computational intelligence for human activity recognition. Therefore, it encourages the submission of state-of-the-art research in HAR, as well as fundamental research relevant to the subject. Topics of interest include (but are not limited to) the following subject categories:
1.IoT Sensor based HAR using Computational Intelligence
2.New algorithms for HAR, including machine learning and deep learning
3.Sensor fusion for HAR
4.Action/gesture detection for autonomous vehicle
5.Action/gesture detection for human-machine interaction
6.Multisensor action recognition
7.Anomaly detection in surveillance videos
8.Wearable HAR applications in the medical field
9.Applications of neural networks in HAR
Zhengyi Chai (Leading Guest Editor)
Special Issue on Computational Intelligence for Human Activity Recognition