Recent advancements in automated welding processes have been remarkable, with vision sensing technologies being essential in boosting their effectiveness. This detailed examination delves into the progress made in vision sensing technologies and their uses in automated joining processes. Advanced imaging technology, paired with cutting-edge processing techniques, delivers non-invasive measurement, exceptional accuracy, and immediate sensing abilities, supplying essential data for the automated management of welding tasks. Nonetheless, acquiring precise depth data continues to be a significant hurdle. Innovative sensing methods, including laser stripes, grid-lines, crosslines, and circular lasers, have significantly enhanced the capability to identify and monitor weld seams, even in intricate shapes and multi-pass scenarios. The incorporation of advanced technologies like artificial intelligence and machine learning, featuring support vector machines and hidden Markov models, has significantly improved the precision and effectiveness of identifying weld types and tracking seams. Although visual detection has shown remarkable advancements, alternative sensing methods, like sound and spectral analysis, contribute essential insights for a thorough grasp of the welding procedure. Future advancements are anticipated to concentrate on boosting depth perception abilities, refining 3D reconstruction precision, and deepening the integration of AI-powered decision-making systems. The integration of advanced vision technologies with AI offers an exciting outlook for automated joining processes, resulting in systems that are more efficient, precise, and adaptable, capable of tackling intricate challenges and enhancing overall production efficiency.