Weld Signal Denoising Algorithm for Welds Based on ICEEMDAN-ICA

With the application of visual sensor in the field of welding seam identification, the welding seam identification technology based on linear array charge coupled devices（CCD） has attracted wide attention due to its advantages such as low cost and good real-time performance, but its accuracy is often closely related to the filtering technology. Therefore, the weld signal denoising algorithm based on ICEEMDAN-ICA was proposed. Firstly, the weld grey image was acquired by using the linear CCD first, and then multiple source of weld signals was obtained by by improved complementary ensemble empirical mode decomposition with adaptive noise and independent component analysis algorithm,the noise signal was identified and eliminated through the fuzzy entropy threshold criterion, finally the weld filtering signal was reconstructed. The experimental results show that this algorithm can remove the high frequency noise of welding signal and retain the effective information.