Many wood identification technologies have been developed or improved in recent years, including attempts to compare results between technologies. The utility of such comparisons is greatly reduced when the species tested with each technology are different and when performance metrics are not calculated or presented in the same way. FPL’s XyloTron was used to develop a species-level computer vision model and was presented along with a side-by-side comparison for species- and genus-level identification of the 10 species of Meliaceae studied by other workers using DART mass spectrometry.
The species-level accuracies of the XyloTron and the DART mass spectrometry models are comparable. However, the genus-level accuracy of the XyloTron model is higher than that of the DART mass spectrometry model. Given that not all wood identification problems are species-level problems, genus-level prediction metrics can be important for real-world adoption of a technology.
Each XyloTron unit costs less than $2000, requires minimal user expertise to operate, is designed to be operated in the field, and after initial acquisition requires only electricity (e.g. the battery power of the laptop, for field use). A mass spectrometer, by comparison, is restricted to a single indoor location, costs approximately $250,000, and requires an operator with specialized expertise to develop an identification result.