本文已被：浏览 1267次 下载 574次
投稿时间：2021-01-06
投稿时间：2021-01-06
中文摘要: 为准确快速提取不同坡口的焊缝特征信息，采用改进Steger算法在滤波去噪后的二值图像中提取出光条纹中心线。首先，将原图像与Otsu法相结合得到包含所有结构光条纹的矩形区域，利用高斯函数的可分离性将图像的二维高斯卷积进行简化，用一次高斯行卷积和一次高斯列卷积进行替代，从而减少运算量。由实验结果可知，改进后的算法不仅能更加准确地提取焊缝光条纹中心线，而且能够提升运行速度，满足实时性的要求。
Abstract:In order to accurately and quickly extract the weld feature information of different grooves, the improved Steger algorithm is used to extract the laser stripe center line from the filtered binary image.Firstly, the ROI region is obtained by combining the original image with the threshold method.Then, according to the separability of Gaussian, the two-dimensional Gaussian convolution of the image is equivalent to a Gaussian row convolution and a Gaussian column convolution, so as to reduce the amount of computation.The experimental results show that the weld recognition algorithm can accurately extract the weld laser stripe center line, and can improve the processing speed to meet the real-time requirements.
文章编号：20212018 中图分类号：TP249;TP391.7 文献标志码：
基金项目:国家自然科学基金（51405286）；上海市重点实验室电站自动化技术实验室项目（13DZ2273800）。
| 作者 | 单位 | |
| 廉杰 | 上海电力大学 | |
| 吕学勤 | 上海电力大学 | lvxueqin@shiep.edu.cn |
| 顾冬霞 | 国网上海青浦供电公司 | |
| 王敏 | 国网上海市北供电公司 |
引用文本：
廉杰,吕学勤,顾冬霞,等.基于焊缝图像信息的光条纹中心线提取[J].上海电力大学学报,2021,37(2):195-199.
LIAN Jie,LYU Xueqin,GU Dongxia,et al.Extraction of Laser Stripe Center Line Based on Weld Image Information[J].Journal of Shanghai University of Electric Power,2021,37(2):195-199.
廉杰,吕学勤,顾冬霞,等.基于焊缝图像信息的光条纹中心线提取[J].上海电力大学学报,2021,37(2):195-199.
LIAN Jie,LYU Xueqin,GU Dongxia,et al.Extraction of Laser Stripe Center Line Based on Weld Image Information[J].Journal of Shanghai University of Electric Power,2021,37(2):195-199.