DSP transplantation of motion object detection algorithm based on
TMS320DM642
Xiaoling Zhang
1
, Peng Xie
1,a
and Baofeng Zhang
1
1
Tianjin Key Laboratory for Control Theory and Applications in Complicated Systems School
of Electrical Engineering, Tianjin University of Technology, Tianjin, 300384 China
a
76033913@qq.com
Keywords: TMS320DM642, algorithm transplantation, Simulink, Real-time Workshop
Abstract. To build motion object detection system based on DSP (Digital Signal Processor), the
motion object detection algorithm by existing is transplanted to the DSP development environment
based on adopted DSP hardware system.DSP migration process of the moving object detection
algorithm is expounded. First of all, Real-time Workshop tool module in Simulink to establish CCS
project file which can be identified by TMS320DM642, then testis processor in the loop (PIL) to
verify its correctness. Create system model by Simulink instead of the traditional programming to
create DSP project files shorten the development cycle and improve the enforceability of the
program. Theresults show that the project files generated by the Simulink can meet the
requirements in terms of complete moving object detection.
Introduction
With the rapid economic development, the increasing progress of science the development of the
society will be derivative a lot of different problems. Therefore, it is vital significance for good
system that detect motion target and video tracking and monitoring ensure social stability and
people's life safety. Therefore, it has certain significance, for in-depth study of moving objects
detection technology. In order to build this system that it is based on DSP platform's moving object
detection system need to transplant the moving target detection algorithm that exist PC to DSP.
DSP transplant process
The algorithm of DSP code generation and migration is based on frame difference method. The
overall migration process is shown in Fig.1.
Fig.1. DSP code migration process
Applied Mechanics and Materials Vol. 577 (2014) pp 790-793
© (2014) Trans Tech Publications, Switzerland
doi:10.4028/www.scientific.net/AMM.577.790
All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of TTP,
www.ttp.net. (ID: 123.150.255.202-20/06/14,09:22:17)