基于格林函数边缘检测的焊缝信息获取 .pdfVIP

学兔兔 2010年第2期 煤 炭 工 程 基于格林函数边缘检测的焊缝信息获取 赵娜娜，黄友锐，涂 帅 (安徽理工大学 电气与信息工程学院，安徽淮南 232001) 摘 要：格林函数边缘检测算法受噪声和阈值干扰较大，针对这一问题，采用各向异性扩散 方程对图像进行去噪处理，并用改进的粒子群算法优化指数熵来确定高低阈值，将改进的格林函 数边缘检测算法引入焊缝图像的处理过程，有效解决了边缘检测技术在抗噪声能力和精确定位能 力间的矛盾。仿真结果表明，改进的算法检测的边缘细、连贯，受噪声的干扰小，将改进的算法 应用于焊缝图像处理起到了很好的效果。 关键词：粒子群算法；指数熵；格林函数；各向异性扩散；焊缝图像 中图分类号：TP391 文献标识码：B 文章编号：1671～0959(2010)02-0087-03 Acquisition of welded seam information base on Green function edge detection ZHAO Na—na．HUANG You—rui．TU Shuai (School of Electrical and Information Engineering，Anhui University of Science and Technology，Huainan 232001，China) Abstract：The green S function edges detection algorithm could be highly interfered by the noise and threshold value． According to the problem，anisotropic diffusion equation was applied to the noise treatment．The improved particle swarm algorithm was applied to optimize the exponent entropy for the determination of the low and high threshold values．The improved green S function edge detection algorithm was introduced to the processing of the seam images and could be effectively to sole the contradiction between the anti noise capacity and accurate positioning capacity for the edge detection technology．The simulation results showed that the edges detected with the improved algorithm was fine and coherence and the noise interference was low．The improved algorithm was applied to the seam image process with good results． Keywords：particle swarm algorithm；exponent entropy；greens function；anisotropic diffusion；welded seam image 子边缘检测方法 ，很难抵抗噪声的干扰，改进的格林函 1 概 述 数边缘检测算法可以有效地