J4

• 计算机科学 • 上一篇    下一篇

嵌入式系统有限资源下的代码移植和优化

赵 颢, 周春光, 吉 斌, 陈媛媛, 于哲舟   

  1. 吉林大学 计算机科学与技术学院, 长春 130012
  • 收稿日期:2006-12-25 修回日期:1900-01-01 出版日期:2007-11-26 发布日期:2007-11-26
  • 通讯作者: 于哲舟

Code Porting and Optimization of Embedded Systemwith Limited Resource

ZHAO Hao, ZHOU Chunguang, JI Bin, CHEN Yuanyuan, YU Zhezhou   

  1. College of Computer Science and Technology, Jilin University, Changchun 130012, China
  • Received:2006-12-25 Revised:1900-01-01 Online:2007-11-26 Published:2007-11-26
  • Contact: YU Zhezhou

摘要: 给出一种基于ADI Blackfin系列数字信号处理器的人像考勤系统设计方案. 在此基础上, 归纳了嵌入式有限资源限制下人像识别算法的实现策略,解决了算法移植过程中从C++到C的转换问题, 并提出了通过修改连接描述文件重新配置存储资源的方法, 从而缓解了大数据量与芯片上有限存储资源之间的矛盾. 测试结果表明, 移植后人像识别算法的运行结果和速度都满足使用要求.

关键词: 移植, 优化, 嵌入式系统, 人像识别

Abstract: An attendance system was designed based on the ADI Blackfin digital signal processors. It summarizesthe code porting methods under the limited resource for embedded system, which resolves the problem to convert C++ to C language; it modifies the linker description files to reconfigure the memory map, which resolves the conflict between the large numbers of data and the limited resource. The test shows the porting results are correct, and the executing speed meets the requirements of application.

Key words: port, optimization, embedded system, face recognition

中图分类号: 

  • TP277