Artificial intelligence and computer vision offer opportunities to automate industrial inspection and quality control. This project develops a bolt inspection and counting system using a webcam and an image classification model trained with Google’s Teachable Machine. As bolts move along a conveyor belt, the webcam captures images that the model classifies according to predefined bolt categories or condition labels. The system also counts bolts passing through the inspection area, integrating condition inspection and quantity monitoring in a single workflow. Its performance will depend on factors such as lighting, camera position, variations in bolt appearance, and the size and quality of the training dataset. The project aims to demonstrate the feasibility of using an accessible computer vision platform for real-time component inspection and counting, with the potential to reduce manual inspection effort and support more efficient production monitoring.
Keywords
computer vision, bolt inspection, object classification, component counting, Teachable Machine, conveyor belt