System Designation and Optimization in Real-Time Identification of Lane Boundary Based on DM642

Abstract:

The hardware and programming of lane boundary identification system, which based on the hardware platform of DM642, designed. The system complete the functions of acquiring image data, optimizing acquired data by ant algorithm, comparing value by the objective function, exporting image of lane boundary and so on. For the sake of improving the real time capability further, optimize code by way of methods, which include the optimization of compiler, intrinsic function, packaging data, using pipeline technique, and assembly code, giving specific optimization methods and real time capability of algorithm affected by each process. As a result the identification time of system reduced from 101.75 ms to 20.55ms, improved the system of real time capability effectively, laying a good foundation for industry.

• Periodical: Advanced Materials Research (Volumes 433-440)
• Pages: 2571-2577
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.433-440.2571
• Online since: January 2012
• Authors: Yu Qin Zhao, Run Shen Zhang, Yong Kun Li, Jun Jie Zang
• Keywords: Compiler Optimization, DM642, Lane Boundary Identification, Pipeline Technique

References

[1] Jungcr, Kelbercr. A robust linear-parabolic model for lane following[C]., In: 17th Brazilian Symposium on the Computer Graphics and Image Processing, 2004，pp.72-79.

DOI: 10.1109/sibgra.2004.1352945

[2] A. Broggi, S. Cattani. An agent based evolutionary approach to path detection for off-road vehicle guidance[J]., Pattern Recognition Letters, Volume 27, Issue 11, pp.1164-1173, August (2006).

DOI: 10.1016/j.patrec.2005.07.014

[3] Gregorr, Lutzelerm, Pellkofer Mela1. EMS-Vision: a perceptual system for autonomous vehicles[J]., IEEE Transactions on Intelligent Transportation Systems. pp.48-59, 2002: 3(1).

DOI: 10.1109/6979.994795

[4] C. R. Jung, C. R. Kelber. Lane following and lane departure using a linear-parabolic model[J]., Image and Vision Computing, pp.1192-2005, 13(23).

DOI: 10.1016/j.imavis.2005.07.018

[5] Y. Bing, Z. Weigong, G. Zongyang. An Embeded Hardware Platform of Lane Departure Warning System Based on DSP[J]., p.28, (2009).

[6] L. Huifang, W. Fei, H. Peikun. TMS320C6000 Series DSPs Theory and Application [M]., Beijing: Electronic Industry Press.

[7] W. Botao. Method of Detecting Anteior Lane And Vehicles of Intelligent Vehicle in Complex Condition[D]., Qin Huangdao: Yan Shan University, (2008).

[8] M. Lei, W. Botao, W. Liandong. Means of Setting The Region of Interesting in Identifying Bent Lane[J]., Journal of Highway And Transportation Research And Development, pp.134-138, 2009, 26(1).

[9] Y. Xiang, Z. Xiaowei, P. Zeyou, Software Optimization in Real-Time Tracking System of Moving Object Based TMS320DM642[J]., Ordnance Industry Automation. 2009, 28.

[10] H. Fei, H. Chunhai, L. Wei. TMS320C6000 Series DSP Development and Application of Skills[M]., China Electric Power Press.

[11] Texas Instruments TM320C6000 Optimizing Compiler user's Guide [Z]. Literature Number: SPRU187O, (2008).

[12] Texas Instruments TMS320C6000 Assembly Language Tools User's Guide[Z]. Literature Number: SPRU186Q, (2008).