ZHANG Li, LIU Bin, GUI Junguo, et al. Location and recognition method of digital tube of digital display instrument under complex environment[J]. Industry and Mine Automation, 2018, 44(4): 85-89. doi: 10.13272/j.issn.1671-251x.2017100053
Citation: ZHANG Li, LIU Bin, GUI Junguo, et al. Location and recognition method of digital tube of digital display instrument under complex environment[J]. Industry and Mine Automation, 2018, 44(4): 85-89. doi: 10.13272/j.issn.1671-251x.2017100053

Location and recognition method of digital tube of digital display instrument under complex environment

doi: 10.13272/j.issn.1671-251x.2017100053
  • Publish Date: 2018-04-10
  • For problem that digital tube of digital display instrument is difficult to locate and accurately recognize in robot inspection system of mine equipment high-voltage distribution room, a location and recognition method of digital tube of digital display instrument under complex environment was proposed. In aspect of digital tube location, feature graph and mark-based watershed algorithm are used to roughly locate digital tube. Then binary image is obtained by use of Otsu threshold method and mark-based watershed algorithm and split by projection, so as to get digital tube character. Non-digital tube area is excluded by local binary pattern feature and support vector machine classification algorithm, which can improve location accuracy of digital tube. In aspect of digital tube character recognition, self-adaptive Canny algorithm is used to extract edge of digital tube character. Then Radon transform is adopted to estimate tilt angle of digital tube character, and filter template is used to remove noise. Finally, threading method is introduced to recognize digital tube character. The experimental results show that the method has high location and recognition accuracy, which can adapt to different lighting, digital tube size and tilt angle.

     

  • loading
  • 加载中

Catalog

    通讯作者: 陈斌, bchen63@163.com
    • 1. 

      沈阳化工大学材料科学与工程学院 沈阳 110142

    1. 本站搜索
    2. 百度学术搜索
    3. 万方数据库搜索
    4. CNKI搜索

    Article Metrics

    Article views (21) PDF downloads(5) Cited by()
    Proportional views
    Related

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return