作者：黄坚1, 刘桂雄1, 王小辉2
作者单位：1. 华南理工大学机械与汽车工程学院, 广东 广州 510640;
2. 广州市光机电技术研究院, 广东 广州 510663
关键词：模式识别与装置;多处理器系统;视觉检测系统;关联处理器调度算法
摘要：
针对滤光片表面缺陷视觉检测系统中在线检测实时性需求对检测速度要求较高,研究一种有效利用可用硬件资源并行处理实时工作提高处理速度的调度优化策略。基于AOE图对滤光片表面缺陷视觉检测系统进行任务级分析,优化事件、活动拓扑关系与任务间冗余的数据相关性、资源相关性,建立并行任务模型;采用关联处理器调度算法(arbitrary processor affinities,APAs)进行并行多处理器调度,指定任务只能被某个处理器集合执行,将期限紧迫、缓存敏感的任务限制在单一处理器,提高资源利用率,改进检测系统实时性。试验结果表明:在尺寸为1.20 mm1.20 mm、2628个滤光片组成滤光片面板上,采用多处理器调度可使检测速度极大提升,采用APAs调度算法后,平均缺陷识别完成时间为常规检测系统时间的36.5%,可以满足在线实时要求,证明应用多处理器调度方法,可以极大提升检测仪器实时性能的有效性。
Multi-processor scheduling of vision detecting system for filter surface defect
HUANG Jian1, LIU Guixiong1, WANG Xiaohui2
1. School of Mechanical and Automotive Engineering, South China University of Technology, Guangzhou 510640, China;
2. Guangzhou Research Institute of O-M-E Technology, Guangzhou 510663, China
Abstract: Online real-time detection is required in vision detecting system for filter surface defect. This paper discusses about a scheduling optimization strategy for utilizing available hardware resources to process real-time work in parallel so as to improve the processing. An AOE graph was used to perform task-level analysis on the abovementioned vision detecting system, optimize the data and resource dependency between events, topological relation of activities and inter-task redundancy, and establish a parallel task model. An arbitrary processor affinities(APAs) was employed for parallel multiprocessor scheduling. The designated task could only be executed by a processor set, i.e., the tasks with imminent deadline and sensitive caches were confined on a single processor to enhance the level of resources utilization and improve the timeliness of detecting system. The results show that, on the panel with 26×28 optical filters(1.20 mm×1.20 mm), multiprocessor scheduling can greatly increase the detection speed. With the APAs algorithm, the average time in identifying defects is 36.5% that of the routine detection system. It has been proven that the multiprocessor scheduling can largely enhance the effectiveness f real-time performance of testing instruments.
Keywords: pattern recognition and instrument;multi-processing system;vision detecting system;APAs
2015, 41(10): 90-93 收稿日期: 2015-02-10;收到修改稿日期: 2015-04-29
基金项目: 广东省产学研项目(01562080172294053)
作者简介: 黄坚(1990-),男,广东揭阳市人,硕士研究生,专业方向为先进传感器技术与应用。
参考文献
[1] Liu Y, Yu F H. Automatic inspection system of surface defects on optical IR-CUT filter based on machine vision[J]. Optics and Lasers in Engineering,2014,55(1):243-257.
[2] Nippolainen E, Sidorov I, Kamshilin A A. Detection of surface defects by means of dynamic speckles[C]//Proc. SPIE 8413, Speckle 2012: V International Conference on Speckle Metrology. SPIE,2012.
[3] Tsapaev, Alexei P, Kretinin, et al. Image segmentation methods in problems of surface defect detection[J]. Computer Optics,2012,36(3):448-452.
[4] Wang X H, Wu J F, Liu G X. Feature selection of surface defects in optical Filter[C]//Proceedings of 2014 International Conference on Computer Science and Applications,2014.
[5] 卫红,曹一鸣,王小辉,等. 光通信滤光片外观特性及缺陷的视觉检测[J]. 中国测试,2014(4):30-32.
[6] Jaghoori, Mohammad M. From nonpreemptive to preemptive scheduling: From single-processor to multi-processor[C]//Proceedings of the ACM Symposium on Applied Computing. SAC,2011.
[7] Salamy H, Ramanujam J. A framework for task scheduling and memory partitioning for multi-processor system-on-chip[C]//High Performance Embedded Architectures and Compilers-Fourth International Conference. Berlin:Springer,2009.
[8] Lele A, Kumar A, Rohith H M, et al. Exact schedulability analysis of global multi-processor IUF scheduling using symbolic model checking[C]//5th International Conference on Industrial and Information Systems.IEEE, 2010.
[9] 付梦瑶. 基于联合信息熵特征选择的滤光片外观缺陷检测研究[D]. 广州:华南理工大学,2014.
[10] 吴俊芳,刘桂雄. 图像边缘过渡区的数学表征方法研究[C]//中国仪器仪表学会.2010中国仪器仪表学术、产业大会(论文集1).中国仪器仪表学会,2010.
[11] Milanez T, Collange S. Thread scheduling and memory coalescing for dynamic vectorization of SPMD workloads[J]. Parallel Computing,2014,40(9):548-558.
[12] Gujarati A, Cerqueira F, Brandenburg B. Multiprocessor real-time scheduling with arbitrary processor affinities: from practice to theory[J]. Real-Time Systems,2015,51(4);440-483.
