MAG molten pool edge detection algorithm based on a fusion of dark channel prior dehazing and image enhancement

Abstract

Metal active gas (MAG) welding is one of the widely applied welding techniques using argon and carbon dioxide as shielding gas. In response to the problem of welding halo and drag shadow during the image acquisition process of it, which makes it difficult to accurately extract the contour of the molten pool, this paper proposes a molten pool edge detection method that combines dark channel prior dehazing (DCPD) and improved single scale Retinex image enhancement algorithm. This method overcomes the problem of excessive edge noise in the original molten pool image and the difficulty in feature extraction caused by the dark part of the molten pool after DCPD processing. Through comparative experiments and ablation experiments, it has been shown that the algorithm proposed in this paper has significantly improved the enhancement effect and feature extraction effect, extracting accurate and complete molten pool contours.

References

[[1]] Xiao H, Song J L, Liu X L, et al.. Feature extraction of 2219 aluminum alloy tungsten helium arc welding molten pool image based on morphological algorithms. Aerospace materials technology, 2019, 49(1): 78-81 [J]

[[2]] Wen H, Zeng J, Bian Z, et al.. Study of weld pool monitoring system based on spatial filtering. Journal of manufacturing processes, 2022, 4: 76 [J]

[[3]] Xu Z H. . Research on image analysis and welding quality prediction of robot welding pool based on deep learning, 2022 Guangdong Guangdong Polytechnic Normal University 000170 [D]

[[4]] Gao Y, Zhong P, Tang X, et al.. Feature extraction of laser welding pool image and application in welding quality identification. IEEE access, 2021, PP(99): 1-1 [J]

[[5]] Duan M R. . Online monitoring of welding quality of stainless steel GTAW in spent fuel pool based on deep learning, 2021 Harbin Harbin Institute of Technology 001765 [D]

[[6]] Marc-André N, Johann F G, Dieter L, et al.. Automatic melt pool recognition in X-ray radiography images from laser-molten Al alloy. Journal of materials research and technology, 2022, 21: 3502-3513, J]

[[7]] Kawahito Y, Matsumoto N, Abe Y, et al.. Relationship of laser absorption to keyhole behavior in high power fiber laser welding of stainless steel and aluminum alloy. Journal of materials processing technology, 2011, 211(10): 1563-1568, J]

[[8]] Gao X, Xiang J, Mhafez K, et al.. Restoration and characteristic analysis of X-ray images of molten pool during laser deep penetration welding. Transactions of the China welding institution, 2013, 34(2): 1-4 [J]

[[9]] Miao L G, Xing F, Shi J J, et al.. Method for extracting the size of laser directed energy deposition melt pool based on edge gradient search. Materials introduction, 2024, 2: 1-15 [J]

[[10]] Zhang J K, Liu H, Ying X Q, et al.. Coordinated underwater dark channel prior for artifact removal of challenging image enhancement. Optoelectronics letters, 2023, 19(07): 416-424, J]

[[11]] Wang Z S, Shi Y H, Hong X B, et al.. Weld pool and keyhole geometric feature extraction in K-TIG welding with a gradual gap based on an improved HDR algorithm. Journal of manufacturing processes, 2022, 73: 409-427, J]

[[12]] Kundu R. A novel technique for texture and edge preservation using bilateral filter. National academy science letters, 2021, 44(2): 117-124, J]

[[13]] He K, Sun J, Tang X. Single image haze removal using dark channel prior. IEEE transactions on pattern analysis & machine intelligence, 2011, 33(12): 2341-2353, J]

[[14]] Jobson D J, Rahman Z U, Woodell G A. Properties and performance of a center/surround retinex. IEEE transactions on image processing, 1997, 6(3): 451-462, J]

[[15]] Zhu X Y, Zeng J D, Song X J, et al.. Molten pool contour segmentation in complex scenes based on improved Clahe algorithm. Hot working technology, 2024, 13: 62-67+74 [J]

[[16]] YUAN Y H, CHEN X L, WANG J D. Object-contextual representations for semantic segmentation[EB/OL]. (2019-09-24) [2023-06-26]. https://arxiv.org/abs/1909.11065v2.