生产线上线缆字符的扫描处理方法

Scanning Processing Method for Image of Cable Characters on Production Lines

针对工业环境下得到的线缆字符图像噪声大、光照不均、清晰度低等所导致的图像处理效果不理想问题,本文提出了一种专门用于线缆字符图像的处理技术。通过采集线缆上的字符图像,对其进行边缘检测以及直线检测分割,将图像中的线缆部分完整分割出来。在此基础上通过去改进光照算法及文本定位分离具体字符块,减小处理区域。最后用二值化处理,得到完整清晰的字符二值图像。实验证明,此方法效果良好,可以有效处理线缆字符图像。

In industrial environment, cable character image always had the problems like noise, uneven illumination and fuzzy. These problems make image treatment efficiency isn＇t good. So, this paper put forward an special processing techniques for image of cable characters. By acquiring cable character image, and using edge detection, line detection segmentation to get the cable portion from whole image. Then, using the improved de-illumination algorithm, text location to locate and separate the character image blocks, it can help to reduce processing area. Finally using the binarization to get the full and clear character binary image. The experiment proving that this way has a good result, can process cable character image efficiently.