Classifying Anomalies on Wood Surface Using Artificial Intelligence

Phua, Yue Xin (2025) Classifying Anomalies on Wood Surface Using Artificial Intelligence. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text PHUA_YUE_XIN_FULL TEXT.pdf Restricted to Registered users only Download (1MB)

Abstract

This project aims to develop an AI-based two-stage wood defect detection system using VGG16 for anomaly classification and YOLOv8 for defect localization. The system is designed to automate the identification and classification of surface defects such as cracks, dents, and gesso lines on wood surfaces, enhancing both speed and accuracy in industrial quality control. The first stage utilizes VGG16, a pre-trained deep learning model, to classify images into normal and potentially defective categories, reducing the workload for the second stage. The second stage employs YOLOv8, an advanced real-time object detection model, to identify and localize specific defects in the wood. The system’s performance will be evaluated based on key metrics including accuracy, precision, recall, F1-score, and mAP, with a focus on ensuring fast inference suitable for real-time applications. This project addresses the need for automated, accurate, and efficient defect detection in wood manufacturing, contributing to improved product quality and reduced waste in production processes. Through fine-tuning pre-trained models and evaluating their performance, this study demonstrates the potential of combining VGG16 and YOLOv8 for practical, industrial defect detection.