19 December 2017 A comparison between skeleton and bounding box models for falling direction recognition
Author Affiliations +
Proceedings Volume 10613, 2017 International Conference on Robotics and Machine Vision; 1061305 (2017) https://doi.org/10.1117/12.2300760
Event: Second International Conference on Robotics and Machine Vision, 2017, Kitakyushu, Japan
Abstract
Falling is an injury that can lead to a serious medical condition in every range of the age of people. However, in the case of elderly, the risk of serious injury is much higher. Due to the fact that one way of preventing serious injury is to treat the fallen person as soon as possible, several works attempted to implement different algorithms to recognize the fall. Our work compares the performance of two models based on features extraction: (i) Body joint data (Skeleton Data) which are the joint’s positions in 3 axes and (ii) Bounding box (Box-size Data) covering all body joints. Machine learning algorithms that were chosen are Decision Tree (DT), Naïve Bayes (NB), K-nearest neighbors (KNN), Linear discriminant analysis (LDA), Voting Classification (VC), and Gradient boosting (GB). The results illustrate that the models trained with Skeleton data are performed far better than those trained with Box-size data (with an average accuracy of 94-81% and 80-75%, respectively). KNN shows the best performance in both Body joint model and Bounding box model. In conclusion, KNN with Body joint model performs the best among the others.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Lalita Narupiyakul, Lalita Narupiyakul, Nitikorn Srisrisawang, Nitikorn Srisrisawang, "A comparison between skeleton and bounding box models for falling direction recognition", Proc. SPIE 10613, 2017 International Conference on Robotics and Machine Vision, 1061305 (19 December 2017); doi: 10.1117/12.2300760; https://doi.org/10.1117/12.2300760
PROCEEDINGS
7 PAGES


SHARE
RELATED CONTENT

Graph optimized Laplacian eigenmaps for face recognition
Proceedings of SPIE (February 07 2015)
Multiclass kernel-based feature extraction
Proceedings of SPIE (March 11 2002)
Effective discriminative TCM-KNN for incremental learning
Proceedings of SPIE (October 29 2009)
Recognition of free-form shapes using spherical SOFMs
Proceedings of SPIE (December 26 2001)
Research on the consistency of LVQ classifier
Proceedings of SPIE (August 20 2010)

Back to Top