摘要
带钢宽度测量需要对带钢图像进行分析判定带钢边缘。一般带钢总是带有划痕、压印、表面夹杂等缺陷,导致边缘难于确定。针对这一问题,提出了一种基于最大类间方差法的带钢边缘判定方法。首先将二维钢板图像简化为一维线阵图像,然后对一维线阵图像进行梯度计算和非极大值抑制,查找出每个灰度值的跳跃位置,最后利用最大类间方差法计算出带钢边缘判定参数,对每个灰度值跳变点进行判定,此方法在带钢边缘判定中取得了良好的效果。
The determination of t he edge of a steel strip needs analyzing its image in measuring the width of the steel strip. Normally, such defects as scratches, imprints, surface inclusions inevitably exist on the surface of a strip steel. As a result it is difficult to determine the edge. For this problem, the determination of the edge of a steel strip based on the OTSU algorithm was proposed. Firstly, the two-dimensional image was simplified into the one-dimensional linear image. Then the gradient calculation and non-maximal suppression for the one-dimensional linear image were carried out in order to locate the skip position for each gray value. Finally the parameters for determination of the edge of a steel strip were calculated by the OTSU algorithm, which can be used to determine the skip points for each grey value. Good results were achieved in determining the edge of a steel strip.
出处
《鞍钢技术》
CAS
2016年第5期31-35,44,共6页
Angang Technology
