基于改进Edlines算法的坡口边缘识别研究  

Research on bevel edge recognition based on improved Edlines algorithm

作　　者：段万政 郭波 曾瑄[1] 徐辉 DUAN Wanzheng;GUO Bo;ZENG Xuan;XU Hui(Jiangxi Provincial Key Laboratory of Precision Drive and Control,Nanchang Institute of Technology,Nanchang 330099,China)

机构地区：[1]南昌工程学院南昌市焊接机器人与智能化技术重点试验室,江西南昌330099

出　　处：《南昌工程学院学报》2021年第1期77-82,共6页Journal of Nanchang Institute of Technology

基　　金：江西省教育厅科学技术研究项目(GJJ180935,GJJ201912);江西省科技厅重点研发计划一般项目(20192BBE50015);南昌工程学院研究生创新项目(YJSCX202010).

摘　　要：为了提高焊缝识别的速度和准确性,提出了一种基于改进Edlines坡口边缘识别算法。首先对图像进行ROI特征区域提取,并对ROI图像进行一系列的预处理,然后对坡口边缘提取;改进了Edlines算法搜索方向和判别方式,用Cannylines算法和Edlines算法提取结果作比对,结果表明提出的改进算法识别速度更快且只提取出焊缝坡口线;最后通过MIG焊、MAG焊中的焊接图像进行边缘识别,验证了算法的适应性。In order to improve the speed and accuracy of weld identification, a groove edge recognition algorithm based on improved Edlines is proposed.Firstly, the ROI feature region of the image is extracted, and a series of preprocessing is carried out on the ROI image.Then, the edge of the groove is extracted, and the search direction and discrimination method of the Edlines algorithm are improved.The results of Cannylines algorithm and Edlines algorithm were compared.The results show that the proposed algorithm is faster in recognition and only the weld groove line is extracted.Finally, the edge of groove is extracted by the improved algorithm.The welding image edge recognition verifies the adaptability of the algorithm.

关 键 词：EDlines 坡口识别 ROI 图像处理 

分 类 号：TG409[金属学及工艺—焊接]