The current state-of-the-art for field wood identification to combat illegal logging relies on experienced practitioners using hand lenses, specialized identification keys, atlases of woods, and field manuals. Accumulation of this expertise is time-consuming and access to training is relatively rare compared to the international demand for field wood identification. A reliable, consistent and cost effective field screening method is necessary for effective global scale enforcement of international treaties such as the Convention on the International Trade in Endagered Species (CITES) or national laws (e.g. the US Lacey Act) governing timber trade and imports. We present highly effective computer vision classification models, based on deep convolutional neural networks, trained via transfer learning, to identify the woods of 10 neotropical species in the family Meliaceae, including CITES-listed Swietenia macrophylla, Swietenia mahagoni, Cedrela fissilis, and Cedrela odorata. We build and evaluate models to classify the 10 woods at the species and genus levels, with image-level model accuracy ranging from 87.4 to 97.5%, with the strongest performance by the genus-level model. Misclassified images are attributed to classes consistent with traditional wood anatomical results, and our species-level accuracy greatly exceeds the resolution of traditional wood identification. The end-to-end trained image classifiers that we present discriminate the woods based on digital images of the transverse surface of solid wood blocks, which are surfaces and images that can be prepared and captured in the field. Hence this work represents a strong proof-of-concept for using computer vision and convolutional neural networks to develop practical models for field screening timber and wood products to combat illegal logging.

Pre-visual detection of crop disease is critical for food security. Field-based spectroscopic remote sensing offers a method to enable timely detection, but still requires appropriate instrumentation and testing. Soybean plants were spectrally measured throughout a growing season to assess the capacity of leaf and canopy level spectral measurements to detect non-visual foliage symptoms induced by Fusarium virguliforme (Fv, which causes sudden death syndrome). Canopy reflectance measurements were made using the Piccolo Doppio dual field-of-view, two-spectrometer (400 to 1630 nm) system on a tractor. Leaf level measurements were obtained, in different plots, using a handheld spectrometer (400 to 2500 nm). Partial least squares discriminant analysis (PLSDA) was applied to the spectroscopic data to discriminate between Fv-inoculated and control plants. Canopy and leaf spectral data allowed identification of Fv infection, prior to visual symptoms, with classification accuracy of 88% and 91% for calibration, 79% and 87% for cross-validation, and 82% and 92% for validation, respectively. Differences in wavelengths important to prediction by canopy vs. leaf data confirm that there are different bases for accurate predictions among methods. Partial least square regression (PLSR) was used on a late-stage canopy level data to predict soybean seed yield, with calibration, cross-validation and validation R² values 0.71, 0.59 and 0.62 (p < 0.01), respectively, and validation root mean square error of 0.31 t·ha⁻¹. Spectral data from the tractor mounted system are thus sensitive to the expression of Fv root infection at canopy scale prior to canopy symptoms, suggesting such systems may be effective for precision agricultural research and management.

Forests, estimated to contain two thirds of the world’s biodiversity, face existential threats due to illegal logging and land conversion. Efforts to combat illegal logging and to support sustainable value chains are hampered by a critical lack of affordable and scalable technologies for field-level inspection of wood and wood products. To meet this need we present the XyloTron, a complete, self-contained, multi-illumination, field-deployable, open-source platform for field imaging and identification of forest products at the macroscopic scale. The XyloTron platform integrates an imaging system built with off-the-shelf components, flexible illumination options with visible and UV light sources, software for camera control, and deep learning models for identification. We demonstrate the capabilities of the XyloTron platform with example applications for automatic wood and charcoal identification using visible light and human-mediated wood identification based on ultra-violet illumination and discuss applications in field imaging, metrology, and material characterization of other substrates.

Illegal logging is a major threat to forests in Peru, in the Amazon more broadly, and in the tropics globally. In Peru alone, more than two thirds of logging concessions showed unauthorized tree harvesting in natural protected areas and indigenous territories, and in 2016 more than half of exported lumber was of illegal origin. To help combat illegal logging and support legal timber trade in Peru we trained a convolutional neural network using transfer learning on images obtained from specimens in six xylaria using the open source, field-deployable XyloTron platform, for the classification of 228 Peruvian species into 24 anatomically informed and contextually relevant classes. The trained models achieved accuracies of 97% for five-fold cross validation, and 86.5 and 92.4% for top-1 and top-2 classification, respectively, on unique independent specimens from a xylarium that did not contribute training data. These results are the first multi-site, multi-user, multi-system-instantiation study for a national scale, computer vision wood identification system evaluated on independent scientific wood specimens. We demonstrate system readiness for evaluation in real-world field screening scenarios using this accurate, affordable, and scalable technology for monitoring, incentivizing, and monetizing legal and sustainable wood value chains.

Availability of and access to wood identification expertise or technology is a critical component for the design and implementation of practical, enforceable strategies for effective promotion, monitoring and incentivisation of sustainable practices and conservation efforts in the forest products value chain. To address this need in the context of the multi-billion-dollar North American wood products industry 22-class, image-based, deep learning models for the macroscopic identification of North American diffuse porous hardwoods were trained for deployment on the open-source, field-deployable XyloTron platform using transverse surface images of specimens from three different xylaria and evaluated on specimens from a fourth xylarium that did not contribute training data. Analysis of the model performance, in the context of the anatomy of the woods considered, demonstrates immediate readiness of the technology developed herein for field testing in a human-in-the-loop monitoring scenario. Also proposed are strategies for training, evaluating, and advancing the state-of-the-art for developing an expansive, continental scale model for all the North American hardwoods.