论文部分内容阅读
图像特征提取尤其是焊缝中心的提取是图像处理与分析中最重要的内容之一。现有的提取方法存在精度低、抗噪性能差、处理速度慢等特点。针对这些问题,本文提出了一种通过求取平均值来提取焊缝中心线的快速有效的算法。该算法采用逐行搜索焊缝左右两个边界点,求取两边界点坐标的平均值作为该道焊缝中心点的坐标。实验证明,该算法在去除噪声、保持焊缝中心连通性以及处理速度方面都取得了不错的效果,并且运算量小,适合焊缝图像的实时处理,能满足实际应用的需要。
Image feature extraction, especially the extraction of the weld center is one of the most important content in image processing and analysis. The existing extraction methods have the characteristics of low precision, poor anti-noise performance and slow processing speed. In response to these problems, this paper presents a fast and effective algorithm for extracting the center line of weld by averaging. The algorithm uses the line-by-line search for the two boundary points around the weld seam to find the average of the coordinates of the two boundary points as the coordinates of the center point of the seam weld. Experiments show that the algorithm has achieved good results in removing noise, maintaining the connectivity of the weld center and processing speed, and the calculation is small, suitable for real-time processing of weld images, to meet the practical application needs.