期刊文献+

面向图像转置和分块处理的一种高效内存访问策略 被引量:3

An Efficient Memory Access Strategy for Transposition and Block Operation in Image Processing
下载PDF
导出
摘要 图像转置和图像分块处理在目前图像处理过程中是经常用到的两种操作.对于实时图像处理,图像转置和图像分块的效率直接影响到实时性,而图像转置和图像分块的效率和内存访问效率密切相关.而访存效率与计算机体系结构、存储器结构和实际操作策略有直接的关系.根据存储器的读写特性提出一种分段存储的高效内存访问策略,通过理论分析得出合理的分段长度,即图像转置时分段长度的选择与DDR2 SDRAM的型号有关;而图像分块时分段长度的选择则与数据块的大小有关.结合工程实际应用还推导了普适的总线地址与存储器物理地址之间的映射,同时给出了一种硬件实现方式,针对不同的应用,仅需要替换地址映射模块,具有一定的通用性和扩展性.分段存储方法已经在SAR实时成像压缩系统中得到有效的验证. Image transposition and image block processing are common operations in image processing. The efficiencies of image transposition and image block processing, which are closely related to memory access efficiency, directly affect the real-time of image processing, while memory access efficiency has direct relationship with computer architecture, memory structure and operational strategy. A segmented storage strategy for efficient memory accessing is proposed in this paper. The segmented storage is based on the read/write characteristics of memory, and the proper segmented length is calculated by theoretic analysis. When the image is transposed, the segmented length is related to the type of DDR2 SDRAM. When the image is partitioned, the segmented length is related to the size of image block. Combined with engineering application, universal mapping between bus address and memory address is deduced in this paper. Meanwhile, a hardware module is given to implement DDR2 SDRAM controller. It is only necessary to update address mapping module for different applications. So the segmented storage strategy has certain universality and expansibility, and is verified in the SAR real-time image process system.
出处 《计算机研究与发展》 EI CSCD 北大核心 2013年第1期188-196,共9页 Journal of Computer Research and Development
关键词 图像转置 图像分块 内存访问 分段存储 地址映射 image transposition image block processing memory access segmented storage address mapping
  • 相关文献

参考文献16

  • 1张冠杰,张涛,张欢阳,王贞松,张守宏.实时合成孔径雷达成像中的专用快速CTM算法[J].西安电子科技大学学报,2006,33(1):11-14. 被引量:4
  • 2谢应科,张涛,韩承德.实时SAR成像系统中矩阵转置的设计和实现[J].计算机研究与发展,2003,40(1):6-11. 被引量:17
  • 3lzumi H, Sasaki K, Nakajima IK, et al. An efficient technique for corner turn in SAR image reconstruction by improving cache access [C] //Proc of Int Parallel and Distributed Processing Symp(IPDPS 2002). Piscataway, NJ: IEEE, 2002: 3-8.
  • 4卢世祥,韩松,王岩飞.合成孔径雷达实时成像转置存储器的两页式结构与实现[J].电子与信息学报,2005,27(8):1226-1228. 被引量:9
  • 5Ren Gang, Han Jizhong, Han Chengde. CTM multiprocessor: Solution for bottleneck of SAR [C] //Proc of the 5th Int Conf on Signal Processing Proc (WCCC ICSP 2000). Beijing: Publishing House of Electronics Industry, 2000, 1915-1920.
  • 6Portnoff M R. An efficient method for transposing large matrices and its application to separable processing of two dimensional signals [J]. IEEE Trans on Image Processing, 1993, 2(1): 122-124.
  • 7Portnoff M R. An efficient parallel-processing method for transposing large matrices in place [J]. IEEE Trans on Image Processing, 1999, 8(9):1265-1275.
  • 8A1 Na'mneh R, Pan W D, Adhami R. Communication efficient adaptive matrix transpose algorithm for FFT on symmetric muhiprocessors [C] //Proc of the 37th Southeastern Symp on System Theory. Piscataway, NJ: IEEE, 2005:312-315.
  • 9AI Na'mneh R, Pan W D, Yoo Seong Moo. Efficient adaptive algorithms for transposing small and large matrices on symmetric multiprocessors [J]. lnformatica, 2006, 17 (4) : 535-550.
  • 10Zhou Qihai, l.i Yan. Isomorphic new parallel division methods and parallel algorithms for giant matrix transpose [J]. Journal of Computers, 2010, 5(5): 169-177.

二级参考文献20

  • 1谢应科 史云涛 等.实时高分辨率SAR成像系统中CTM的设计[J].计算机研究与发展,2000,37:195-199.
  • 2Panchanathan S.Universal Architecture for Matrix Transposition[J].IEE Proceedings E,1992,139(5):387-392.
  • 3Tierno J A,Kudva P.Asynchronous Transpose-matrix Architectures[A].IEEE International Conference on Computer Design:VLSI in Computers and Processors[C].Austin:IEEE,1997.423-428
  • 4Marino F,Swartzlander E E Jr.Parallel Implementation of Multidimensional Transforms Without Interprocessor Communication[J].IEEE Trans on Computer,1999,48(9):951-961.
  • 5Chatterjee S,Sen S.Cache-efficient Matrix Transposition[A].High Performance Computer Architecture[C].Toulous:IEEE,2000.195-205.
  • 6Kim K,Beerl P A.MSB-Controlled Inversion Coding for a Low Power Matrix Transposer[J].Electronics Letters,1999,35(17):1 434-1 435.
  • 7张涛.SAR实时成像处理技术[R].北京:中国科学院计算技术研究所,2003..
  • 8Ren Gang, Han Jizhong, Han Chengde. CTM on multiprocessor:Solution for bottleneck of SAR. International Conference of Signal Processing, Beijing, 2000:1917 - 1918.
  • 9Micron Incorporated.Small-Outline DDR SDRAM DIMM Data Book, 2003.
  • 10Altera Incorporated.Stratix Device Hand Book, 2003.

共引文献18

同被引文献36

  • 1卢世祥,韩松,王岩飞.合成孔径雷达实时成像转置存储器的两页式结构与实现[J].电子与信息学报,2005,27(8):1226-1228. 被引量:9
  • 2李早社,禹卫东,汪亮,郑小双.基于SDRAM的星载SAR星上实时成像转置存储器[J].信号处理,2007,23(3):433-436. 被引量:5
  • 3P.Soille,王小鹏.形态学图像分析原理与应用[M].北京:清华大学出版社,2008.
  • 4孙燮华.数字图像处理原理与方法[M].北京:机械工业出版社,2012.
  • 5ZHAOYindi, LIU Shanlei. Feature - basedGeometric Reg- istration of High Spatial Resolution Satellite Imagery[ C ]. 2009 Urban Remote Sensing Joint Event, 2009.
  • 6SAEED K,TABEDZKI M, RYBNIK M, et al. K3M: A Universal Algorithm for Image Skeletonization and a Re- view of Thinning Techniques [ J ]. Applied Mathematics and Computer Science, 2010,20 ( 2 ) :317 - 335.
  • 7MICHAEL J JONES, PAUL VIOLA. Fast multi - view face detection[ C]. IEEE Conference on Computer Vision and Pattern Recognition, 2003.
  • 8OSADCHY M, MILLER M, LECUN Y. Synergistic face detection and pose estimation with energy - based models [ C ] Submitted to NIPS, 2004.
  • 9ABDENOUR HADID, MATH PIETIKAINEN, TIMO AHONEN. A Discriminative Feature Space for Detecting and Recognizing faces[ C ]. Proceeding of the IEEE Com- puter Society Conference on Computer Vision and Pattern Recognition, 2004.
  • 10王林泉.SAR实时成像系统中的压缩和矩阵转置研究[D].西安:西安电子科技大学.2015.

引证文献3

二级引证文献3

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