面向多目标透镜的连通域标记检测算法

doi:10.16180/j.cnki.issn1007-7820.2020.04.009

电子科技 ›› 2020, Vol. 33 ›› Issue (4): 50-54.doi: 10.16180/j.cnki.issn1007-7820.2020.04.009

面向多目标透镜的连通域标记检测算法

谢智烜¹,姚红兵²,范宁¹,陈枫¹

^1. 江苏大学机械工程学院,江苏镇江 212000
^2. 河海大学理学院,江苏南京 210098

收稿日期:2019-03-01 出版日期:2020-04-15 发布日期:2020-04-23
作者简介:谢智烜(1991-),男,硕士研究生。研究方向:图像处理,目标识别等。|姚红兵(1976-),男,教授。研究方向:现代光电测试,激光与物质相互作用机理以及检测技术。
基金资助:
国家自然科学基金(51775253);国家自然科学基金(61505071);江苏省自然科学基金(BK20150526);江苏大学引进人才科研启动项目(15JDG036);江苏大学引进人才科研启动项目(15JDG037)

Connected Domain Label Detection Algorithm for Multi-target Lens

XIE Zhixuan¹,YAO Hongbing²,FAN Ning¹,CHEN Feng¹

^1. School of Mechanical Engineering,Jiangsu University,Zhenjiang 212000,China
^2. School of Science,Hohai University,Nanjing 210098,China

Received:2019-03-01 Online:2020-04-15 Published:2020-04-23
Supported by:
National Natural Science Foundation of China(51775253);National Natural Science Foundation of China(61505071);Jiangsu Natural Science Foundation(BK20150526);Jiangsu University's Initiation Project of Introducing Talented Personnel for Scientific Research(15JDG036);Jiangsu University's Initiation Project of Introducing Talented Personnel for Scientific Research(15JDG037)

摘要/Abstract

摘要：

文中提出了一种适合FPGA实现的多目标连通域标记算法。该算法通过初次扫描时在图像的边框处做特殊的标记值,提取出环形具有内外双层轮廓的目标内部区域,进而将紧挨在一起的目标分离开以实现多目标的检测。该方案采用单次逐像素扫描的方法,对图像进行一次逐像素扫描即可完成连通域的标记和等价表的整理。实验结果表明,对一幅大小为M×N的图像,在不计算等价表递归时间的情况下,经过M×N×2个时钟周期后可识别出多个目标,且得到了每个目标对应区域的面积和外接矩形等参数。

关键词: 图像处理, 多目标图像, 连通域标记, FPGA, 机器视觉, 二值图像

Abstract:

A multi-target connected domain labeling suitable algorithm for FPGA was proposed in the study. By making a special mark value at the edge of the image during the first scanning, the algorithm extracted ring inner region with inner and outer double-layer contours object to separate the close objects for multi-target detection. The scheme adopted a single pixel-by-pixel scanning method to perform one scanning of the image, and could accomplish the mark of connected regions and the sorting of equivalent tables. Experimental results showed that for an image with the size of M×N, without calculating the recursive time of the equivalent table, multiple targets could be identified after M×N×2 clock cycles. In addition, parameters including the area and the outer rectangle corresponding to each target were finally obtained.

Key words: image processing, multi-target image, connected domain markers, FPGA, machine vision, binary image

中图分类号:

TP391

谢智烜,姚红兵,范宁,陈枫. 面向多目标透镜的连通域标记检测算法[J]. 电子科技, 2020, 33(4): 50-54.

XIE Zhixuan,YAO Hongbing,FAN Ning,CHEN Feng. Connected Domain Label Detection Algorithm for Multi-target Lens[J]. Electronic Science and Technology, 2020, 33(4): 50-54.

图/表 10

图1

图2

图3

表1

图4

图5

图6

表2

表3

图7

参考文献 16

[1]	Telljohann A. Introduction to building a machine vision inspection,handbook of machine vision[M]. Darmstadt: Wiley Press, 2007.
[2]	刘紫燕, 唐虎, 冯亮 . 基于FPGA的机器视觉技术视频跟踪系统实现[J]. 电子科技, 2018,31(9):41-49.
	Liu Ziyan, Tang Hu, Feng Liang . Realization of video tracking system for machine vision technology based on FPGA[J]. Electronic Science and Technology, 2018,31(9):41-49.
[3]	Andersson T, Thurley M J, Carlson J E . A machine vision system for estimation of size distributions by weight of limestone particles[J]. Minerals Engineering, 2016,25(1):38-46. doi: 10.1016/j.mineng.2011.10.001
[4]	Jeong J, Lee G, Lee M , et al. Asingle-pass connected component labeler without label merging period[J]. Journal of Signal Processing Systems, 2016,84(2):211-223. doi: 10.1007/s11265-015-1048-7
[5]	张奎, 陈兆学 . 一种基于游程码的二值图像孔洞标记和表达方法[J]. 光学技术, 2017,43(4):364-368.
	Zhang Kui, Chen Zhaoxue . Holes labeling & expressing method in binary images based on run-length code[J]. Optical Technique, 2017,43(4):364-368.
[6]	张国和, 徐快, 段国栋 , 等. 一种适于硬件实现的快速连通域标记算法[J]. 西安交通大学学报, 2018,52(8):95-101.
	Zhang Guohe, Xu Kuai, Duan Guodong , et al. Afast labeling algorithm of connected components applicable for hardware implementation[J]. Journal of Xi'an Jiaotong University, 2018,52(8):95-101.
[7]	Gu Q, Takaki T, Ishii I . Fast FPGA-based multiobject feature extraction[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2013,23(1):30-45. doi: 10.1109/TCSVT.2012.2202195
[8]	戴华东, 胡谋法 . 基于连通域标记的目标检测算法设计与实现[J]. 现代电子技术, 2015,38(20):71-74.
	Dai Huadong, Hu Moufa . Design and implementation of target detection algorithm based on connected domain labeling[J]. Modern Electronics Technique, 2015,38(20):71-74.
[9]	马毅超, 陶亚凡, 吴龙胜 , 等. 基于FPGA的二值图片连通区域标记算法[J]. 微电子学与计算机, 2017,34(1):119-122.
	Ma Yichao, Tao Yafan, Wu Longsheng , et al. FPGA-based binary connected component detection and labeling algorithm[J]. Microelectronics and Computer, 2017,34(1):119-122.
[10]	陈小光, 王琳, 汪周武 . 一种基于连通域分析的卡片指纹图像分割算法[J]. 北方工业大学学报, 2015,27(1):1-6.
	Chen Xiaoguang, Wang Lin, Wang Zhouwu . Afast andefficient algorithm for fingerprint segmentation based on connected analysis[J]. Journal of North China University of Technology, 2015,27(1):1-6.
[11]	于潇宇, 郭玉波 . 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):1-11
	Yu Xiaoyu, Guo Yubo . Real-time point feature extraction based on connected components labeling and distributed computing[J]. Acta Optica Sinica, 2015,35(2):1-11.
[12]	李艳婷, 王水鱼 . 视频图像中动态目标检测系统的FPGA硬件设计[J]. 计算机系统应用, 2017,26(3):119-125.
	Li Yanting, Wang Shuiyu . FPGA hardware design of dynamic target detection system in video image[J]. Computer Systemsand Applications, 2017,26(3):119-125.
[13]	王凯, 黄山, 赵瑜 , 等. 面向图像目标提取的改进连通域标记算法[J]. 计算机工程与设计, 2014,35(7):2439-2441.
	Wang Kai, Huang Shan, Zhao Yu , et al. Improved algorithm of connected component labeling for image targets extraction[J]. Computer Engineering and Design, 2014,35(7):2439-2441.
[14]	赵晓, 何立风, 姚斌 , 等. 一种基于连通域标记的纸病检测算法[J]. 中国造纸学报, 2018,33(2):51-56.
	Zhao Xiao, He Lifeng, Yao Bin , et al. Apaper disease detection algorithm based on connected region marking[J]. Transactions of China Pulp and Paper, 2018,33(2):51-56.
[15]	姑丽祖热·吐尔逊, 尤努斯·艾沙, 吐尔根·依布拉音 , 等. 连通域结合重叠度的维吾尔文档图像文字切分[J]. 计算机工程与设计, 2016,37(7):1892-1897.
	Gulzira Tursun, Yunus Aysa, Tuergin Yibulayin , et al. Combination of connected regions and overlapping degree based Uyghr document image text segmentation[J]. Computer Engineering and Design, 2016,37(7):1892-1897.
[16]	桑红石, 赵慧, 尚社 . 多值图像连通域标记ASIC结构设计[J]. 小型微型计算机系统, 2008,29(1):180-184.
	Sang Hongshi, Zhao Hui, Shang She . Architecture design of ASIC used for connected components labeling of muilt-value image[J]. Journal of Chinese Computer Systems, 2008,29(1):180-184.

数据位	标记参数
0~10	标记值
11~21	X_max
22~32	X_min
33~43	Y_max
44~54	Y_min
55~69	面积

临时标记值	X_max	X_min	Y_max	Y_min	面积
3	269	149	805	694	9 775
89	433	312	770	656	9 861
142	524	429	858	785	5 035
217	687	593	871	793	5 150

面向多目标透镜的连通域标记检测算法

Connected Domain Label Detection Algorithm for Multi-target Lens

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 16

相关文章 15

Metrics

本文评价

推荐阅读 10

[1]	宗圣康,程建鹏,张西良. 基于光斑位置的起重机轨道高度差自动检测方法[J]. 电子科技, 2022, 35(1): 21-28.
[2]	马鼎,吴定祥,唐立军. 环形陶瓷金属涂层缺陷检测方法研究[J]. 电子科技, 2021, 34(6): 28-33.
[3]	杨凌霄,王振营,刘群坡,王高伟,杨彦超. 基于H-S直方图的苹果着色率检测算法研究[J]. 电子科技, 2021, 34(5): 13-17.
[4]	宋章明,贺慧勇,黄跃俊. 扁平漆包线表面缺陷在线检测研究[J]. 电子科技, 2021, 34(5): 72-78.
[5]	闫超,孙占全,田恩刚,赵杨洋,范小燕. 基于深度学习的医学图像分割技术研究进展[J]. 电子科技, 2021, 34(2): 7-11.
[6]	晏子杰,王京梅,陈卓,刘宇. 软件可编程的FPGA网络测量引擎技术实现[J]. 电子科技, 2021, 34(2): 27-32.
[7]	陈文佑,章伟,史晓帆,宋芳. 一种改进ORB特征匹配的半稠密三维重建ORB-SLAM算法[J]. 电子科技, 2021, 34(12): 62-67.
[8]	史伟忠,曹卫卫,范彦铭,董佳筠,陈纾,肖昊. 基于FPGA的深空图像实时边缘检测算法与实现[J]. 电子科技, 2020, 33(5): 45-49.
[9]	章天昊,李郝林. 一种基于高斯拟合法的激光光束中心提取方法[J]. 电子科技, 2020, 33(5): 72-76.
[10]	秦兴,高晓琪,陈滨. 基于压缩卷积神经网络的图像超分辨率算法[J]. 电子科技, 2020, 33(5): 1-8.
[11]	詹雁,张娟. 一种结构感知损失的域适应深度估计方法[J]. 电子科技, 2020, 33(12): 12-16.
[12]	赵能武,路宏敏,徐桃,胡宽,何川侠,孟晓姣. 基于AD9910的周界入侵探测系统信号源设计[J]. 电子科技, 2020, 33(1): 1-5.
[13]	段铁,黄焱,汪洋. 载波同步中数字锁相环的重构设计[J]. 电子科技, 2019, 32(8): 41-45.
[14]	骆英,吴强,秦云. GSC框架下的最小均方超声波束形成算法[J]. 电子科技, 2019, 32(2): 37-41.
[15]	王家敏,杨青慧,张怀武. 9 kHz1.4 GHz高精度快速随机跳频DDS频率合成器[J]. 电子科技, 2019, 32(12): 27-32.