Silhouette Orientation Volumes for Efficient Fall Detection in Depth Videos

Akagunduz E., Aslan M., Sengur A., Wang H., Ince M. C.

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, vol.21, no.3, pp.756-763, 2017 (SCI-Expanded) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 21 Issue: 3
  • Publication Date: 2017
  • Doi Number: 10.1109/jbhi.2016.2570300
  • Journal Name: IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.756-763
  • Keywords: Bag of words, fall detection, naive Bayes classifier, SDU-fall dataset, shape matching, silhouette orientation volume, weizmann action dataset
  • Middle East Technical University Affiliated: No

Abstract

A novel method to detect human falls in depth videos is presented in this paper. A fast and robust shape sequence descriptor, namely the Silhouette Orientation Volume (SOV), is used to represent actions and classify falls. The SOV descriptor provides high classification accuracy even with a combination of simple associated models, such as Bag-of-Words and the Naive Bayes classifier. Experiments on the public SDU-Fall dataset show that this new approach achieves up to 91.89% fall detection accuracy with a single- view depth camera. The classification rate is about 5% higher than the results reported in the literature. An overall accuracy of 89.63% was obtained for the six-class action recognition, which is about 25% higher than the state of the art. Moreover, a perfect silhouette-based action recognition rate of 100% is achieved on the Weizmann action dataset.