基于点目标连通域标记的实时特征提取及其分布式运算

在高速运动目标的视觉测量中，高分辨率、高帧频的图像序列带来了大量待处理数据，如何快速地从这些数据中识别合作目标并提取其特征信息，成为高速视觉测量中的难题。对此，针对高速相机每个时钟周期多像素并行传输的数据特点，提出一种基于多维金字塔的硬件加速处理结构，实现连通域的全局搜索与标记，并根据标记结果完成对应特征的实时提取。在现场可编程逻辑门阵列中，通过金字塔结构的二维处理节点阵列与多标签特征统计的一维处理阵列，将数据流的高密度运算均衡分布于各运算节点，结合流水线并行，将视觉系统中占据较高耗时的全局搜索与标记过程在图像传输中同步实现。通过功能验证与实时性分析，该特征提取的分布式运算结构可在CameraLink接口85 MHz的时钟频率下，实现680 Mpixel/s的数据流实时处理，可作为预处理部分应用于高速视觉测量系统。

Abstract

In the vision measurement for high speed moving target, the image sequence with high resolution and high frame rate brings a lot of datas. How to recognize the cooperation target quickly and extract the feature information in real-time is becoming a challenge in vision measurement system. In each clock cycle of high speed camera, multi- pixel is transmitted in parallel. In this situation, a hardware accelerating structure based on multi- dimensional pyramid is proposed, the task of global searching and connected components labeling is realized, the features of connected components are extracted real- timely based on the labeling results. In field programmable gate array, with the pyramid structure formed by two- dimensional processing element (PE) array and multi-label feature statistical structure formed by one-dimensional PE array, the high density computing process of data flow is distributed into each PE node equally. Combining with the parallel pipeline, the global searching and labeling process which takes a remarkable time- consuming in ordinary vision system is accomplished synchronously with image transmission. It is verified that the distributed computing structure for feature extraction can deal with real- time data processing at the flow rate of 680 Mpixel/s at the frequency of 85 MHz with CameraLink interface. As a preprocess part, it can be used in high-speed vision measurement system.

参考文献

[1] 郭寅, 刘常杰, 邾继贵, 等. 高速列车动态位姿测量方法及校准技术研究[J]. 光电子·激光, 2013, 34(1): 112-118.

Guo Yin, Liu Changjie, Zhu Jigui, et al.. Research on measurement and calibration of dynamic posture of high-speed train [J]. Journal of Optoelectronics·Laser, 2013, 34(1): 112-118.

[2] G Johansson. Visual motion perception [J]. Sci Am, 1975, 232(6): 76-88.

[3] F Zhu, S Qin. A detection algorithm for maneuvering IR point target and its performance evaluation [C]. 2006 IEEE/ASME International Conference on Mechatronics and Embedded Systems and Applications, 2006: 1-5.

[4] N Kumar, N Kunju, Amod Kumar, et al.. Active marker based kinematic and spatio- temporal gait measurement system using Labview vision [J]. Journal of scientific & Industrial Resarch, 2010, 69: 600-605.

[5] 江洁, 苗准, 张广军. 点结构光动态姿态角测量系统[J]. 红外与激光工程, 2010, 39(3): 532-536.

Jiang Jie, Miao Zhun, Zhang Guangjun. Dynamic altitude angle measurement system based on dot-structure light [J]. Infrared and Laser Engineering, 2010, 39(3): 532-536.

[6] D Ye, X Yu, G Chen, et al.. Real- time preprocessing for vision measurement system on high- speed cooperative target [C]. International Conference on Mechatronics and Control, 2014.

[7] 于潇宇, 郭玉波, 陈刚, 等. 视觉测量中点目标的实时全局搜索方法[J]. 光学精密工程, 2014, 22(10): 2796-2805.

Yu Xiaoyu, Guo Yubo, Chen Gang, et al.. Real-time global search for point targets in vision measurement[J]. Optics and Precision Engineering, 2014, 22(10): 2796-2805.

[8] C T Johnston, D G Bailey. FPGA implementation of a single pass connected components algorithm [C]. Delta 2008: Fourth IEEE International Symposium on Electronic Design, Test and Applications, 2008, 228-231.

[9] N Ma, D G Bailey, C T Johnston. Optimised single pass connected components analysis [J]. Proceedings of the 2008 International Conference on Field-Programmable Technology, 2008, 185-192.

[10] D G Bailey, C T Johnston, N Ma, et al.. Connected components analysis of streamed images [C]. Proceedings of the 18th International Conference on Field Programmable and Logic Applications, 2008, 679-682.

[11] A W Malik, B Thornberg, M Imran, et al.. Hardware architecture for real-time computation of image component feature descriptors on a FPGA [J]. International Journal of Distributed Sensor Networks, 2014, 815378.

[12] D K Kim, D R Lee, T C Pham, et al.. Real-time component labeling and boundary tracing system based on FPGA [C]. 2007 IEEE International Conference on Robotics and Biomimetics, 2007, 189-194.

[13] 张广军. 机器视觉[M]. 北京: 科学出版社, 2005. 99-102.

Zhang Guangjun. Machine Vision [M]. Beijing: Science Press, 2005. 99-102.

[14] 刘博, 叶东, 陈刚, 等. 火箭喷管运动视觉测试精度的校准与实验[J]. 光学精密工程, 2010, 18(11): 2513-2520.

Liu Bo, Ye Dong, Chen Gang, et al.. Calibration and experiment of vision measurement accuracy for motion of rocket nozzle [J]. Optics and Precision Engineering. 2010, 18(11): 2513-2520.

于潇宇, 郭玉波, 陈刚, 叶东. 基于点目标连通域标记的实时特征提取及其分布式运算[J]. 光学学报, 2015, 35(2): 0210001. Yu Xiaoyu, Guo Yubo, Chen Gang, Ye Dong. Real-Time Point Feature Extraction Based on Connected Components Labeling and Distributed Computing[J]. Acta Optica Sinica, 2015, 35(2): 0210001.

基于点目标连通域标记的实时特征提取及其分布式运算

关于本站 Cookie 的使用提示

全站搜索

基于点目标连通域标记的实时特征提取及其分布式运算

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索