【摘要】 针对带钢板在生产中存在的大边浪和碎浪检测的难点，提出一种基于机器视觉的边浪平直度测量方法。分别在带钢两侧使用高分辨率相机，完成对钢板两侧边浪图像信息的采集；使用辅助光源强化带钢板边浪特征，实现非接触式测量。基于轮廓提取的带钢钢板边浪平直度算法，使用改进K-means聚类图像分割算法对钢板完成分割；引入容差，计算统计平均值，对钢板边浪进行分割；最后对分割后的边浪进行自适应轮廓提取并计算平直度。通过平直度测量仪测量结果与该系统测量结果进行对比试验，结果表明，所提方法平直度测量相对偏差不超过0.25%，表明该算法在平直度测试中具有出色的表现。更多还原

【Abstract】 Aiming at the difficulties of detecting large edge waves and broken waves in the production of strip steel plate, a machine vision-based edge wave straightness measurement method is proposed. A high-resolution camera is used on both sides of the strip to collect the image information of the waves on both sides of the steel plate, and an auxiliary light source is used to strengthen the wave characteristics of the strip to achieve noncontact measurement. Based on the contour extraction algorithm of the strip steel plate edge wave straightness algorithm, using the improved K-means clustering image segmentation algorithm to complete the segmentation of the steel plate; the introduction of tolerance, the calculation of the statistical average value, the segmentation of the steel plate edge wave; and finally, the segmented edge wave adaptive contour extraction and calculation of the straightness. Comparison test is carried out between the measurement results of the straightness measuring instrument and those of the system, and the results show that the relative deviation of the straightness measurement of the proposed method is not more than 0.25%, which demonstrates that the algorithm has excellent performance in straightness testing.更多还原