从图像中快速检测直线的并行算法  

Fast Parallel Algorithm for Detecting Lines from Image

作　　者：虞凡[1] 吴惠思[1] 覃征[1] 杨博[1] 

机构地区：[1]西安交通大学电子与信息工程学院,西安710049

出　　处：《西安交通大学学报》2006年第12期1370-1373,1387,共5页Journal of Xi'an Jiaotong University

基　　金：国家重点基础研究计划资助项目(2004CB719401)

摘　　要：提出了一种在具有可扩展机群体系结构的通用超级并行机环境下的快速直线检测算法.采用最优域划分法将原图像进行N×N等面积划分,并映射到并行系统各节点的局部坐标下进行直线检测.通过先后两次在Hough变换的扫描过程中提前引入阈值作用,可有效减小计算的复杂度.同时证明了,划分后的子图像采用原阈值的1/(2N-1)进行检测,可满足原图像中的直线目标不丢失的要求.实验结果表明,所提算法具备高鲁棒性和强抗噪能力,能有效提升加速比,该比值的最大值可为25.320.A quick lines detection algorithm under the general super parallel system environment with scalable cluster structure is presented. It divides the original image into N × N parts with same area by means of the partition method of optimal domain, and maps each sub image onto the local coordinates of the nodes of the parallel system for detecting lines. Two thresholds are preintroduced to the two different scanning periods of Hough transform to simplify calculation. It is proved that the detection with 1/（2N-1） of the original threshold on the divided sub image can satisfy the demands that are no target loss in the original image. Experiment result indicates that the algorithm effectively reduces the computation complexity and raises the accelerating ratio up to 25. 320, while being highly robust and noise avoidable.

关 键 词：并行算法 图像处理 直线检测 

分 类 号：TP302[自动化与计算机技术—计算机系统结构]