Abstract: At present, the detection of the depth of bar rolling passes in China relies entirely on manual measurement. Different personnel may result in different results, which may lead to significant deviations in roll gap settings. To achieve the purpose of automatic measurement of hole depth mentioned above, this paper studies a method for detecting hole depth in bar racks based on standard comparative magnetic rods as specimens, parallel light source optical imaging as a means, and image processing technology, as well as related mathematical modeling and algorithms. The above method can balance the convenience and accuracy of on-site roll changing operation and production preparation, and has the characteristics of high accuracy and easy operation. On site experiments have shown that the hole depth detection method based on image processing can achieve a measurement accuracy of less than 0.2 mm and is relatively stable, which has certain practical application value for guiding actual industrial production.

Key words: Auto bar rolling; Roll gap measurement; Image processing

摘要： 目前国内棒材轧制孔型深度的检测，全部依靠人工卡量，不同的人员测量，会导致不 同的结果，进而可能会导致辊缝设置存在较大的偏差。为实现上述孔型深度的自动测量目的， 文章研究了一种以标准对比磁棒为标本、平行光源光学成像为手段、图像处理技术为基础的 棒材机架孔型深度检测方法，以及相关的数学建模及算法。上述方法能够兼顾现场换辊操作 及生产准备的便利性和精度，具有精度高、操作方便的特点。现场实验表明，以图像处理为 基础的孔型深度检测方法，其实测精度可以达到 0.2 mm 以内，且比较稳定，对指导实际工业 生产具有一定的实际应用价值。

关键词: 汽车用棒材轧制；辊缝测量；图像处理