摘要:
由于MAG焊外部干扰大,用CCD采集的MAG焊熔池区图像不够清晰。因此,在进行进一步处理比如边缘检测之前,需要进行必要的图像增强和去噪等预处理操作,这样才能获得更加清晰的图像边缘信息。针对以上问题,采用超分辨率技术对MAG焊熔池图像进行处理和分析。在此基础上,提出了基于小波分形插值的超分辨率图像重建方法。结果表明,该方法相对于基于传统的小波双线性插值的超分辨率方法,能有效提高MAG焊图像的分辨率,获得更加清晰的焊缝位置和图像的边缘信息,为后续的焊接质量控制和焊缝跟踪工作提供了很好的基础。
Abstract:
MAG welding pool images captured by charge coupled device (CCD)is not clear due to a serious external interference, which has worse impact on further research work.Therefore, before further processing such as edge detection some image preprocessing operation such as image enhancement and de-noising should be necessary to obtain more clear image edge informatiort.Superresolution technique was used to process and analyze MAG welding pool image aiming at the above problems, and super-resolution image reconstruction methods based on wavelet fractal interpolation was proposed, The experimentd results indicate that the proposed method can effectively improve the image resolution compared with superresolution method based on traditional wavelet Bilinear interpolation, can provide more valuable position and edge information of welding seam, which founds a sound base for the quality control and welding seam tracking
下载:
