Characterization of sawlogs using industrial X-ray and 3D scanning

Sammanfattning: In the Nordic countries, sawlogs are typically sorted upon arrival at the sawmill based on species and dimension. By processing batches of logs with similar size, the sawing process becomes more efficient; the need to change sawing pattern between individual logs is reduced, and the handling of sawn goods is simplified, since the number of different dimensions produced simultaneously decreases. However, since wood is a biological material with great heterogeneity, there will be a large variation in the properties also of boards sawn from logs of similar size. This means that a significant amount of the boards may be carrying unwanted combinations of dimension and grade, so called off-grade products. The problem with off-grade products may be addressed before sawing by the selection of suitable sawing patterns for each log, i.e., using the right logs for the right products. This requires knowledge of the internal quality of the log before sawing. Some information can be obtained from the outer shape measured by an optical three-dimensional (3D) scanner and more detailed information can be obtained using an X-ray log scanner. Today, the use of X-ray log scanners is becoming increasingly common, and most sawmills installing an X-ray scanner already have a 3D scanner present. This raises the question of possible benefits from combining the X-ray and 3D scanning techniques. In this thesis, a method is presented whereby the outer shape of the log measured by a 3D scanner is utilized to estimate the X-ray path lengths through the wood. This converts the X-ray images into green density images of the log, which may in turn be used to calculate quality variables such as heartwood diameter, dry density, moisture content and presence of top rupture. The methods have been tested on Scots pine (Pinus sylvestris L.) and Norway spruce (Picea abies (L.) Karst.) sawlogs using X-ray scanners with one or two measurement directions. The developed methods show a great improvement in precision compared to calculations based on uncompensated X-ray images, and most of the algorithms presented in the thesis have now been implemented in industrial scanner software and are ready for use at the sawmills. This will give the sawmill industry new possibilities to control the production of special products where heartwood diameter and density are important and will lead to less waste and improved profitability for the sawmills.The thesis also describes a method where X-ray scanning is utilized to automatically perform parts of the log grading for payment. This method can improve productivity in the sawmills by remedying a severe bottleneck in the production chain. An authorization of this method for semiautomatic log grading for payment is expected to further increase the industry’s interest in X-ray scanning.