Feature fusion H-ELM based learned features and hand-crafted features for human activity recognition

• Main Authors: AlDahoul, Nouar, Htike, Zaw Zaw
• Format: Article
• Language: English; English; English
• Published: 2018
• Subjects: Q350 Information theory
• Online Access: http://irep.iium.edu.my/69665/25/69665_Feature%20Fusion_%20H-ELM%20based%20Learned_article.pdf; http://irep.iium.edu.my/69665/19/69665_Feature%20Fusion%20H-ELM%20based%20learned%20features%20and%20hand-crafted%20features_scopus.pdf; http://irep.iium.edu.my/69665/26/69665_Feature%20Fusion_%20H-ELM%20based%20Learned_wos.pdf; http://irep.iium.edu.my/69665/; http://thesai.org/Publications/IJACSA
• Institution: Universiti Islam Antarabangsa Malaysia

Recognizing human activities is one of the main goals of human-centered intelligent systems. Smartphone sensors produce a continuous sequence of observations. These observations are noisy, unstructured and high dimensional. Therefore, efficient features have to be extracted in order to perform accurate classification. This paper proposes a combination of Hierarchical and kernel Extreme Learning Machine (HK-ELM) methods to learn features and map them to specific classes in a short time. Moreover, a feature fusion approach is proposed to combine H-ELM based learned features with hand-crafted ones. Our proposed method was found to outperform state-of-the-art in terms of accuracy and training time. It gives accuracy of 97.62 % and takes 3.4 seconds as a training time by using a normal Central Processing Unit (CPU).