激光深熔焊熔池X光图像恢复与特征分析

Restoration and characteristic analysis of X-ray images of molten pool during laser deep penetration welding

X光对于焊件具有很强的穿透性,可以利用其观测和分析激光深熔焊过程焊件内部熔池形态以及匙孔能量传递效果,实现对焊接过程有效地实时监测.在大功率光纤激光焊接304不锈钢过程中,由于X光图像采集过程中的噪声干扰,导致熔池X光图像存在严重退化现象,清晰度不高,识别性较差.为此研究一种图像恢复和增强算法,对熔池X光退化图像进行恢复和增强处理,使其在处理后能够更好地被识别.以盲去卷积法试验估计出熔池X光图像退化的点扩散函数,并以此函数为参数对熔池X光图像进行维纳滤波恢复,再进行灰度变换以增强图像,进而对熔池动态特性进行分析.结果表明,所提出的方法能够有效改善激光焊过程熔池X光图像质量,突出熔池边缘轮廓特征.

Owing to a strong ability to go through the objects,the radiography can be used to observe and analyze the formation of a molten pool inside weldments during laser deep penetration welding.The shape of a molten pool and the thermal transfer effect of laser power through keyhole can be monitored and analyzed in real-time.During a high-power fiber laser welding on plate of 304 stainless steel,a high-speed radiography camera was used to capture the molten pool images.These captured X-ray images were degenerated and obscure due to the disturbance and noises from the welding process and radiography devices.An arithmetic to restore and intensify the X-ray images of molten pool was proposed to make the X-ray images more clear and enhance the ability to recognize the characteristics of images.The blind deconvolution method was used to estimate the point spreading function of X-ray image degeneration,which was applied as a parameter to implement the Wiener filtering of X-ray image of a molten pool.Also,the X-ray image of a molten pool was intensified by the grey level transformation,and the dynamic of a molten pool was analyzed.Results showed that the proposed algorithm of image restoration and intensification can improve the quality of X-ray images effectively and make the contour of a molten pool more distinct.