• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2014, Vol. 36 ›› Issue (04): 751-757.

• 论文 • 上一篇    下一篇

一种新的渣土车车轮带泥检测方法

杨雪1,李志华1,2   

  1. (1.江南大学物联网工程学院轻工过程先进控制教育部重点实验室,江苏 无锡 214122;
    2.物联网应用技术教育部工程研究中心,江苏 无锡 214122)
  • 收稿日期:2012-10-23 修回日期:2013-01-21 出版日期:2014-04-25 发布日期:2014-04-25
  • 基金资助:

    中央高校基本科研业务费专项资金资助项目(JUSRP211A41)

A new detection method of mudattaching dregs
transportation vehicle wheel              

YANG Xue1,LI Zhihua1,2   

  1. (1.Key Laboratory of Advanced Process Control for Light Industry of Ministry of Education,
    School of IoT Engineering,Jiangnan University,Wuxi 214122;
    2.Engineering Research Center of IoT Technology Application of Ministry of Education,Wuxi 214122,China)
  • Received:2012-10-23 Revised:2013-01-21 Online:2014-04-25 Published:2014-04-25

摘要:

提出了一种新的车轮带泥图像检测方法。通过改进直方图方法、在相似度计算中考虑了像素的位置信息,从而提出了直方图交叉相似度计算的新方式;并综合考虑小波变换法在纹理检测上的优势,给出了图像组合相似度的新定义,基于此,提出了基于小波变换提取边缘特征与改进的直方图交叉方法相结合的相似度比较EHSC算法,通过EHSC算法计算组合相似度,并根据相似度阈值来最终选择图像。EHSC算法具有时间开销小、鲁棒性好等特点。通过在真实采样的现场图像上实验,实验结果表明,该方法能有效地进行带泥车轮图像的检测,具有较高的检测效率和比较强的实用性。

关键词: 车轮带泥检测, 边缘检测, 直方图交叉, EHSC算法, 相似度比较

Abstract:

A new method for mudattaching wheel images detection is proposed.By introducing a new definition of combining similarity of images(CSI), Edge Histogram Similarity Comparison(EHSC) algorithm is proposed by combining image edge and histogram features. Firstly, wavelet transformation is used to extract edge feature of images.Thus,wavelet feature similarity between realtime images and reference map is calculated.Then,the similarity between realtime images and reference map is calculated by the improved histogram intersection method.Finally,the images of satisfying threshold condition are selected. The simulation experiment result shows that the method has the advantage of strong robustness,low overhead of time and higher accuracy.It can improve the efficiency of detecting mudattaching wheel images efficiently.Key words:

Key words: wheel mud-attaching detection, edge detection;histogram intersection;EHSC algorithm;similarity comparison