期刊文献+

多值图像连通域标记ASIC结构设计 被引量:4

Architecture Design of ASIC Used for Connected Components Labeling of Multi-value Image
下载PDF
导出
摘要 提出了一种能够实时实现多值图像连通域标记算法的ASIC系统结构.该器件采用基于像素的图像连通域标记算法,通过分析存储器需求,减少了片内存储器容量,以及存储器访问所需要的时钟周期数,使图像标记的总时钟数降为N×M×4,N为图像行数,M为列数.仿真结果表明,ASIC能够满足大部分实时目标识别系统的要求. An architecture of connected components labeling algorithm used for multi-value segmented image labeling is pro- posed in this paper. Algorithms adopted in this architecture is connected components labeling algorithm based on pixel. By ana- lyzing memory requirement, the in-chip memory capacity reduced, and clock cycles required by memory accessing are reduced either. Total clock cycles to process an image is N * M * 4, in which N is the row of the image and the M is the column of the image. Simulation results show that the ASIC based on the architecture can be used in many real-time pattern recognition systems.
出处 《小型微型计算机系统》 CSCD 北大核心 2008年第1期180-184,共5页 Journal of Chinese Computer Systems
基金 空间电子技术国家级重点实验室基金项目(20060126)资助
关键词 连通域标记 多值图像 ASIC 系统结构 connected component labeling multi-value image ASIC architecture
  • 相关文献

参考文献8

  • 1Vikrant Khanna, Gupta P, Hwang C J. Finding connected components in digital Images[C]. In: Proc. International Conference on Information Technology: Coding and Computing, 2001, 652-656.
  • 2Suzuki K, Horiba I, Sugie N. Fast connected-component labeling based on sequential local operations in the course of forward raster scan followed by backward raster scan[C]. In: Proc. 15th International Conference on Pattern Recognition, 2000, Vol. 2, 434-437.
  • 3Yang X D. Design of fast connected components hardware[C]. In: Proc. Computer Vision and Pattern Rection Conference, 1988, 937-944.
  • 4Ranganathan N, Mehrotra R, Subramanmian S. A high speed systolic architecture for labeling connected components in an image[C]. IEEE Transaction on Systems, Man, and Cybernetics, 1995, 25(3): 415-423.
  • 5Rasquinha, Ranganathan N. C3L: a chip for connected component labeling[C]. In: Proc. VLSI Design, 1997, 446-450.
  • 6Nicol C J. A systolic approach for realtime connected component labeling [C]. CVGIP : Image Understanding, 1995,61 (1) : 17- 31.
  • 7Zhang Gui-lin, Li Qiang. A multiple-target image tracker based on real-time labeling[J]. Journal of Huazhong University of Science and Technology, Nature Science, 1994, 22(5), 36-41.
  • 8Sang Hong-shi, Fu Yong. A connected components labeling algorithm for multi-value image that suitable for realization in VLSI[J]. Journal of Huazhong University of Science and Technology, Nature Science, 2005,33(9),5-8.

同被引文献27

  • 1桑红石,傅勇,张天序,刘云生.一种适合硬件实现的多值图像连通域标记算法[J].华中科技大学学报(自然科学版),2005,33(9):5-8. 被引量:5
  • 2冈萨雷斯.数字图像处理[M].阮秋琦,阮宇智,译.2版.北京:电子工业出版社,2007:427.
  • 3FLATT H, BLUME S, HESSEI.BARTH S, et al. A pa-rallel hardware architecture for connected component labeling based on fast label merging[C]//Application-Specific Systems, Architectures and Processors.[S. 1. ]: IEEE, 2008:14,1-149.
  • 4YANG S W, SHEU M H, WU H H, et al. VLSI architecture design for fast parallel lahel assignment in hinary image[C]. Circuits and Systems. [S. 1. ]: IEEE, 2005:2393-2396.
  • 5JOHNSTON C T, BAILEY D G. FPGA implementation of a single pass connected components algorithm [C]//Electronic Design, Test and Applications. [S. 1.]: IEEE, 2008:228-231.
  • 6RASQUINHA A. RANGANATHAN N. C3L: a chip for connected component labeling[C]// VLSI Design. [S. 1. ]. IEEE, 1997:446- 450.
  • 7James R Armstrong,F Gail Gray.VHDL设计表示与综合[M].北京:机械工业出版社,2003.
  • 8Steve Kilts.高级FPGA设计结构、实现和优化[M].北京:机械工业出版社,2009.
  • 9Martin-Herrero. Hybrid object labeling in digital images. [ J ]. Machine Vision and Applications. 2007,18 ( 1 ) : 1-15.
  • 10SUZUKI K, HORIBA I, SUGIE N. Linear time connected component labeling based on sequential local operations. [ J]. Computer Vision and Image Understanding. 2003, 89 (1) :1-23.

引证文献4

二级引证文献25

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部