Automated Chipboard to Edge Band Matching

In the furniture industry, chipboard is a widely used material due to its cost-effectiveness and versatility. However, matching chipboard surfaces with aesthetically and functionally compatible edge bands remains a critical and challenging task, often performed manually based on subjective judgment. This paper presents a novel chipboard edge band matching system that leverages one-shot learning with Siamese networks to automate and optimize this process. The system employs CNN-based Siamese Networks to analyze and quantify the similarity between chipboards and edge bands based on multiple criteria, including color, texture, pattern, and shape. A comprehensive similarity model and a similarity value table are generated, enabling manufacturers to identify the most compatible edge bands for any given chipboard. In addition, this study contributes to the field by introducing an open-source dataset publicly available containing chipboard and edge band samples. This dataset provides a valuable resource for researchers exploring industrial material matching and computer vision applications. The proposed system addresses the limitations of existing methods, offering an efficient, scalable, and objective solution for edge band selection. By bridging the gap between manual processes and advanced automation, this work aims to improve production efficiency, consistency, and aesthetic quality in the furniture manufacturing industry.