| Title | Artificial Intelligence-Enabled Digital Twins for Personalized Healthcare: Architecture, Predictive Analytics, and Implementation Challenges |
| Research Area | Artificial Intelligenc |
| Abstract | Quality inspection is a fundamental component of modern manufacturing because even minor defects
may reduce product reliability, increase material waste, create warranty costs, and negatively
influence customer satisfaction. Conventional quality inspection commonly depends on manual visual
examination or rule-based machine vision. Although these methods remain useful in controlled
production environments, their effectiveness may decline when manufacturing lines operate at high
speed or when defects exhibit substantial variation in size, shape, texture, orientation, and
appearance.
Artificial intelligence, computer vision, and deep learning provide an alternative approach in which
visual inspection systems can automatically learn relevant representations from manufacturing
images. This paper presents a comprehensive framework for computer vision-based automated
quality inspection in smart manufacturing environments. The proposed architecture combines
industrial image acquisition, preprocessing, deep feature extraction, defect classification, object
detection, semantic segmentation, anomaly detection, confidence estimation, edge inference, quality
decision support, and continuous model improvement |
| Keywords | Computer Vision, Deep Learning, Smart Manufacturing, Automated Visual Inspection, Defect Detection, Quality Control, Artificial Intelligence, Anomaly Detection, Industry 4.0, Edge AI. |
| Paper Status | Published |
| Volume | 2 |
| Issue | 5 |
| Published On | 05/10/2026 |
| Published File |
IJSRTD_4896.pdf
|