摘要
为检测连续焊接条件下熔池振荡频率,提出了一种激光视觉测量的熔池振荡检测方法,介绍了激光视觉测量的原理、搭建了试验平台.采集反映熔池表面振荡变化的图像,并进行了裁剪、中值滤波、顶帽变换、二值化处理;编写程序获取熔池振荡的时域信号,进一步通过快速傅里叶变换算法提取了熔池振荡频率.结果表明:该方法能够实现连续焊接条件下熔池振荡频率的检测;顶帽变换图像算法能将整幅图像背景亮度均匀化,为提取完整的激光条纹及熔池振荡频率奠定了基础.
There exists a direct relationship between the weld pool oscillation frequency and states of weld beam penetration. But it is difficult to detect weld pool oscillation. In order to detect the weld pool oscillation frequency during the travelling welding, a novel detecting and sensing method based on laser vision for obtaining the weld pool oscillation frequency was proposed. The principle of the laser vision was introduced and the experiment system to monitor the weld pool oscillation signals was designed. The video ima- ges which can reflect weld pool variation were collected by high-speed cameras. The images acquired were processed by a series of image processing methods. The characteristic frequency was extracted by using FFT. The results show that the top-hat can homogenize the background brightness of the images acquired and the binary images with clear laser stripes can be acquired, laying a foundation for further detection of weld pool oscillation.
作者
石玗
李春凯
顾玉芬
朱明
樊丁
SHI Yu LI Chunkai GUYufen ZHUMing FAN Ding(State Key Laboratory of Advanced Processing and Recycling of Nonferrous Metals, Key Laboratory of Non-Ferrous Metal Alloys and Processing of the Ministry of Education, Lanzhou University of Technology, Lanzhou 730050, Chin)
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2016年第12期1910-1914,共5页
Journal of Shanghai Jiaotong University
基金
国家自然科学基金(51305189)
甘肃省自然科学基金(145RJZA119)
北京石油化工学院光机电装备技术北京市重点实验室开放课题(KF2015-1)资助
关键词
激光视觉
熔池振荡
顶帽变换
图像处理
laser vision
weld pool oscillation
top-hat
image processing