Mechanical Properties in Kinematics Based on Pulse Coupled Neural Network

DOI

10.2991/amcce-15.2015.145How to use a DOI?

Keywords

Aerobics; Movement analysis; Biological vision; Multi core DSP; Low power consumption; Bandwidth

Abstract

In order to improve the efficiency of aerobics movement image analysis, on the basis of biological vision and image processing technology, this paper designs the image motion analysis system with the core of DSP+FPGA circuit and the high-speed external memory allocation. This system is based on the image segmentation and clustering algorithm as the theoretical model to ensure the system having enough computing power on the hardware, at the same time providing sufficient bandwidth for the large amount of image transmission. The test shows that the image processing system has reached the real-time and stability of aerobics movement analysis requirement, further realizing low power consumption and improving the efficiency of image processing algorithm.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Yali Xu
AU  - Kaihua Xu
PY  - 2015/04
DA  - 2015/04
TI  - Mechanical Properties in Kinematics Based on Pulse Coupled Neural Network
BT  - Proceedings of the 2015 International Conference on Automation, Mechanical Control and Computational Engineering
PB  - Atlantis Press
SP  - 1107
EP  - 1112
SN  - 1951-6851
UR  - https://doi.org/10.2991/amcce-15.2015.145
DO  - 10.2991/amcce-15.2015.145
ID  - Xu2015/04
ER  -