A Robotic Smart System to Identify and Classify the Defects in the Manufactured Products

Abstract

In the industrial sector, maintaining product quality is crucial to ensuring customer happiness and lowering production costs. Recently, deep learning techniques have been used to overcome surface-defect detection issues in industrial quality control. In order to effectively and efficiently detect in the manufacturing industries, this research study offers a novel surface product defect detection system developed using the YOLOv5 algorithm. A prototype of the system is implemented using a Raspberry Pi and other components to demonstrate its practical application in the manufacturing industry. The system's effectiveness in accurately detecting surface defects in real-time is shown by experimental findings from the prototype implementation. The YOLOv5 model achieves robust defect detection capabilities through intensive training and fine-tuning, enabling real-time identification of defects in PVC pipes. By highlighting the versatility and flexibility of the training process, the system shows its potential for many manufacturing industries. The proposed system provides an economical and effective method for automating defect identification, hence increasing industrial quality control procedures. It may, therefore, be integrated into current production lines. It has the potential to significantly improve quality control procedures in the manufacturing sector, resulting in better product quality and lower operational costs.