自然场景中基于局部轮廓特征的类圆对象识别方法

doi:10.11772/j.issn.1001-9081.2016.05.1399

计算机应用 ›› 2016, Vol. 36 ›› Issue (5): 1399-1403.DOI: 10.11772/j.issn.1001-9081.2016.05.1399

自然场景中基于局部轮廓特征的类圆对象识别方法

班孝坤¹, 韩军¹, 陆冬明¹, 王万国², 刘俍²

1. 上海大学通信与信息工程学院, 上海 200444;
2. 国网山东省电力公司电力科学研究院国家电网电力机器人技术研究室, 济南 250002

收稿日期:2015-09-01 修回日期:2015-10-19 出版日期:2016-05-10 发布日期:2016-05-09
通讯作者: 韩军
作者简介:班孝坤(1989-),男,安徽宿州人,硕士研究生,主要研究方向:数字图像处理;韩军(1965-),男,河南三峡门人,高级工程师,博士,主要研究方向:图像处理、视频修复;陆冬明(1991-),男,江苏靖江人,硕士研究生,主要研究方向:数字图像处理;王万国(1984-),男,山东聊城人,工程师,硕士,主要研究方向:基于数字图像的变电站和输电线路设备识别、目标跟踪、图像去雾;刘俍(1984-),男,山东济南人,工程师,硕士,主要研究方向:无人机巡检。
基金资助:
2014年国家电网发展项目(169)。

Similar circular object recognition method based on local contour feature in natural scenario

BAN Xiaokun¹, HAN Jun¹, LU Dongming¹, WANG Wanguo², LIU Liang²

1. School of Communication and Information Engineering, Shanghai University, Shanghai 200444, China;
2. Electric Power Robotics Laboratory of State Grid Corporation of China, Shandong Electric Power Research Institute, Jinan Shangdong 250002, China

Received:2015-09-01 Revised:2015-10-19 Online:2016-05-10 Published:2016-05-09
Supported by:
This work is supported by the National Grid Development Project(169).

摘要/Abstract

摘要： 在自然场景下,受背景纹理、遮挡、光线影响,不容易提取到对象完整的轮廓,为此,提出了一种基于局部轮廓特征的识别方法,该局部轮廓特征是由直线和曲线组成的2相邻轮廓片段特征(2AS)。首先,分析相邻片段之间的夹角、片段的长度和弯曲强度,定义2AS的语义模型;接着,依据2AS之间的相对位置关系定义2AS相互关系模型,分别描述对象的2AS特征和2AS之间的相互关系;然后,使用对象模板的2AS的语义模型与测试图像中的2AS特征进行初步匹配,接着依据对象模板的2AS相互关系模型进行精确匹配;最后,根据对象模板的2AS相互关系模型检测到的成组2AS进行重复性聚类,并根据对象模板的2AS相互关系模型对聚类的对象判决。与使用近似直线片段组成的2AS特征算法的对比实验结果证明,该算法对输电线路中均压环部件的识别具有较高的正确率、较低的误检率和漏检率,从而更为有效地识别均压环部件。

关键词: 对象识别, 局部轮廓特征, 相邻片断, 相互关系模型

Abstract: In the natural scenario, it is difficult to extract a complete outline of the object because of background textures, light and occlusion. Therefore an object recognition method based on local contour feature was proposed. Local contour feature of this paper formed by chains of 2-adjacent straight and curve contour segments (2AS). First, the angle of the adjacent segments, the segment length and the bending strength were analyzed, and the semantic model of the 2AS contour feature was defined. Then on the basis of the relative position relation between object's 2AS features, the 2AS mutual relation model was defined. Second, the 2AS semantic model of the object template primarily matched with the 2AS features of the test image, then 2AS mutual relation model of object template accurately matched with the 2AS features of the test image. At last, the pairs of 2AS of detected local contour features were obtained and repeatedly grouped, then grouped objects were verified according to the 2AS mutual relation model of object template. The contrast experiment with the 2AS feature algorithm with similar straight-line chains, the proposed algorithm has higher accuracy, low false positive rate and miss rate in the recognition of grading ring, then the method can more effectively recognize the grading ring.

Key words: object recognition, local contour feature, adjacent segment, mutual relation model

中图分类号:

TN911.73

班孝坤, 韩军, 陆冬明, 王万国, 刘俍. 自然场景中基于局部轮廓特征的类圆对象识别方法[J]. 计算机应用, 2016, 36(5): 1399-1403.

BAN Xiaokun, HAN Jun, LU Dongming, WANG Wanguo, LIU Liang. Similar circular object recognition method based on local contour feature in natural scenario[J]. Journal of Computer Applications, 2016, 36(5): 1399-1403.

参考文献

[1] GHOSH A, PETKOV N. Robustness of shape descriptors to incomplete contour representations[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(11):1793-1804.
[2] BIEDERMAN I, JU G. Surface versus edge-based determinants of visual recognition[J]. Cognitive Psychology, 1988, 20(1):38-64.
[3] RAVISHANKAR S, JAIN A, MITTAL A. Multi-stage contour based detection of deformable objects[C]//Proceedings of the 10th European Conference on Computer Vision. Heidelberg:Springer, 2008:483-496.
[4] SHOTTON J, BLAKE A, CIPOLLA R. Multiscale categorical object recognition using contour fragments[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(7):1270-1281.
[5] FERRARI V, TUYTELAARSTINNE T, van GOOL L. Object detection by contour segment networks[C]//Proceedings of the 9th European Conference on Computer Vision. Heidelberg:Springer, 2006:14-28.
[6] HUET B, HANCOCK E R. Shape recognition from large image libraries by inexact graph matching[J]. Pattern Recognition Letters, 1999, 20(11):1259-1269.
[7] CHANG W, LEE S Y. Description of shape patterns using circular arcs for object detection[J]. IET Computer Vision, 2013, 7(2):90-104.
[8] XU C, KUIPERS B. Object detection using principal contour fragments[C]//Proceedings of the 2011 Canadian Conference on Computer and Robot Vision. Piscataway, NJ:IEEE, 2011:363-370.
[9] LU C E, LATECKI L J, ADLURU N, et al. Shape guided contour grouping with particle filters[C]//Proceedings of the 2009 IEEE 12th International Conference on Computer Vision. Piscataway, NJ:IEEE, 2009:2288-2295.
[10] FERRARI V, FEVRIER L, JURIE F, et al. Groups of adjacent contour segments for object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(1):36-51.
[11] FERRARI V, JURIE F, SCHMID C. Accurate object detection with deformable shape models learnt from images[C]//Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ:IEEE, 2007:1-8.
[12] 朱铭武, 韩军, 路东明, 等. 自然场景中基于局部轮廓特征的对象识别方法[J].计算机工程与应用, 2016, 52(1):162-167.(ZHU M W, HAN J, LU D M, et al. Object recognition method based on local contour feature in natural scenario[J].Computer Engineering and Applications, 2015, 52(1):162-167.)
[13] 王璇, 张帆, 程京. 基于改进最大类间方差法的灰度图像分割[J]. 微计算机信息, 2010, 26(35):206-207.(WANG X, ZHANG F, CHENG J. Improved OTSU method on gray level image segmentation[J]. Control & Automation, 2010, 26(35):206-207.)
[14] 韩军, 张晶晶. 基于感知组织的输电线路结构识别方法[J]. 红外与激光工程, 2013, 42(12):3458-3463.(HAN J, ZHANG J J. Method on recognizing the structure of transmission line based on perceptual organization[J]. Infrared and Laser Engineering, 2013, 42(12):3458-3463.)
[15] NGUYEN T P, DEBLED R I. Decomposition of a curve into arcs and line segments based on dominant point detection[C]//Proceedings of the 17th Scandinavian Conference on Image Analysis. Heidelberg:Springer, 2011:794-805.

自然场景中基于局部轮廓特征的类圆对象识别方法

Similar circular object recognition method based on local contour feature in natural scenario

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

[1]	朱杰, 吴树芳, 谢博鋆, 马丽艳. 基于颜色的压缩层次图像表示方法[J]. 计算机应用, 2017, 37(11): 3238-3243.
[2]	王万国, 张晶晶, 韩军, 刘俍, 朱铭武. 基于无人机图像的输电线断股与异物缺陷检测方法[J]. 计算机应用, 2015, 35(8): 2404-2408.