AI-based quality inspection and defect detection systems are revolutionizing real-time manufacturing environments by enhancing precision, speed, and operational efficiency. This study investigates the deployment of computer vision and deep learning algorithms, particularly convolutional neural networks (CNNs), for automated surface defect detection, dimensional accuracy assessment, and pattern recognition in high-speed production lines. A hybrid edge-cloud architecture is proposed to support low-latency image processing and centralized model optimization. Real-time sensor fusion and adaptive learning techniques are employed to detect anomalies in dynamic industrial settings, significantly reducing false positives and inspection time. Case studies from the automotive and electronics sectors demonstrate over 95% detection accuracy and a 30% improvement in inspection throughput. The paper also discusses integration challenges with legacy systems and proposes a framework using OPC-UA and Industry 4.0 standards. This research validates AI’s role in achieving consistent product quality, minimizing waste, and supporting zero-defect manufacturing strategies in smart factories.
Wai Yie Leong (Mon,) studied this question.