Constructing Vision System for Extracting Profile of Large-Sized Sprocket Products

Along with the development of science and technology, non-contact measurement using image processing techniques has emerged as a popular method for inspection and quality assessment. However, when dealing with large-sized objects, this method encounters challenges in terms of image quality and measurement accuracy. This paper introduces a vision system that utilizes a telecentric lens with a small field of view (FOV) to reconstruct the 2D profiles of large-size sprockets with diameters of up to 150 mm. Instead of capturing the entire object in a single image, each tooth is individually captured to ensure sharpness and high-resolution images. By arranging the obtained sub-profiles next to each other, the complete shape of the object can be reconstructed, enabling high-accuracy measurements and defect detection in the manufacturing process. This also presents a simple application of the vision system to reduce the workload of visual inspection for workers over an extended period.