作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (15): 197-200. doi: 10.3969/j.issn.1000-3428.2011.15.063

• 图形图像处理 • 上一篇    下一篇

多核环境下的图像分割并行算法研究

刘张桥1a,王成良1b,焦晓军1a,2   

  1. (1. 重庆大学 a. 计算机学院;b. 软件学院,重庆 400044;2. 重庆科技学院电气与信息工程学院,重庆 401331)
  • 收稿日期:2011-02-18 出版日期:2011-08-05 发布日期:2011-08-05
  • 作者简介:刘张桥(1983-),男,硕士,主研方向:数字图像处理,多核并行计算;王成良,教授;焦晓军,工程师、硕士
  • 基金资助:
    重庆市科技攻关计划基金资助项目(CSTC, 2009AC2060)

Research of Parallel Algorithm for Image Segmentation Under Multi-core Environment

LIU Zhang-qiao  1a, WANG Cheng-liang  1b, JIAO Xiao-jun  1a,2   

  1. (1a. College of Computer Science; 1b. School of Software Engineering, Chongqing University, Chongqing 400044, China; 2. School of Electrical & Information Engineering, Chongqing University of Science & Technology, Chongqing 401331, China)
  • Received:2011-02-18 Online:2011-08-05 Published:2011-08-05

摘要: 对多核环境下的图像分割并行算法进行研究,在基于正交小波分解的多分辨率图像锥中引入模糊C-均值(FCM)算法,采用OpenMP语言设计P-FCM多核并行模型,并给出该模型的算法实现步骤。在对初始图像数据预处理时,采用矩形块数据分割法进行图像分块,将分块后的子图像数据作为并行运算时的输入数据由主线程分给不同的处理器。实验结果表明,在处理较大图像时,该算法效率较高。

关键词: 图像分割, 多核处理器, OpenMP语言, 模糊C-均值算法

Abstract: This paper studies the parallel algorithm for image segmentation under multi-core environment. It combines the multiresolution image pyramid based on the orthogonal wavelet decomposition with the Fuzzy C-means(FCM) clustering algorithm, and achieves the multi-processor and multi-threaded parallel model of P-FCM with the parallel language of OpenMP. It gives the algorithm implementation of the model. For the original image data preprocessing, the parallel method of rectangular block is used to divide the original image data into sub-rectangular data blocks, which will be used as the input data and assigned to different processor by the mater thread for the parallel computing. Experimental results show that, in dealing with the large size images, this algorithm has high efficiency.

Key words: image segmentation, multi-core processor, OpenMP, Fuzzy C-means(FCM) algorithm

中图分类号: