Screw Defect Sorting system using Data Interpretation

This project presents an automated screw defect detection system using a Raspberry Pi, aimed at enhancing quality control in manufacturing and assembly lines. The system utilizes the Raspberry Pi Camera Module to capture images of screws as they pass through the inspection area. Using image processing techniques implemented in OpenCV, the system analyzes these images to detect common defects such as missing screws, incorrect alignment, damaged threads, and surface deformities. For more complex defect identification, the system optionally integrates a lightweight machine learning model, deployed with TensorFlow Lite, to classify screws as defective or non-defective based on pre-trained visual features. The screw inspection system is powered by the Raspberry Pi’s GPIO pins, which control a servo motor and a gear motor to manage the positioning and movement of screws in real-time. This cost-effective and compact solution provides rapid defect detection, enabling immediate response on the production line, such as rejecting defective screws or alerting operators. Additionally, the system can be extended to include IoT connectivity, allowing remote monitoring and logging of defects for data analysis and preventive maintenance. This approach leverages the affordability and processing capabilities of the Raspberry Pi to create an efficient, scalable solution suitable for industrial automation in small- to medium-scale manufacturing environments.