Raw Wood Fabrication with Computer Vision

Thinned wood, which is removed by pruning and thinning, is not widely used because of its small size and small diameter. Conventional processing tools for building materials cannot be used for thinned wood with small cross-sections because they make the cross-section rectangle. However, using computer vision technology to measure the shape of the wood and a fabrication machine that can adapt to the measurement results makes it possible to process the wood to retain a large cross-section and use it for construction. The developed workflow can be divided into three parts: 1) database to handle wood inventory and shape data, 2) design of the desired shape, and 3) machining. Although it is difficult to reposition a discrete shape of raw wood during processing with the accuracy maintained, the proposed method achieves the repositioning of the wood with high accuracy. This research broadens the scope of raw wood processing and helps to utilize wood that was previously discarded and demonstrates a new processing method for irregular shapes.