Design and Implementation of a YOLOv11-Based Classification System for Composite Boards

This study proposes an automatic classification system for composite wood panels based on an improved YOLOv11 model, aiming to reduce misclassification and labor costs caused by the diverse colors and textures of composite panels. The model integrates channel attention (CA) and salient feature fusion (VFF) techniques to improve the classification accuracy of different composite wood panels. A custom dataset of 15,000 images of 27 synthetic wood panels was collected and annotated. All images were taken under consistent lighting and fixed focus to ensure image consistency. Experimental results show that the improved YOLOv11 model achieves an average precision (mAP) of 99.81% on various composite wood board datasets, and surpasses the performance of the original YOLOv11 algorithm. This innovative technology is expected to not only improve production efficiency and material quality but also significantly reduce testing costs and human resource investment, contributing to the continuous innovation and progress of traditional industries.