If you work in the forestry industry, then you'd better know that you're cutting down the right type of trees. With this is mind, researchers from Finland's Tampere University of Technology have teamed up with colleagues at the Natural Resources Institute Finland, to create a new laser-based tree classification system.
Based on previous research conducted by the university, the technique involves first performing laser scans on known species of trees. This is done to create a 3D point cloud computer model of the tree.
After a sufficient number (about 30) of each tree type is scanned, the models can be analyzed for species-specific characteristics such as the structure of the trees' crowns and trunks, along with the diameter, volume and angle of the branches. A total of 15 parameters are taken into account.
When an unknown species of tree is subsequently scanned, its characteristics can be cross-referenced against those of known species already in the database.
So far, the system has been set up to identify birch, pine and spruce, which it's reportedly able to do with over 95 percent accuracy. The scientists now plan on expanding the database to include a larger number of species.
