Fraud and misrepresentation in retail forest products exceeds U.S. forensic wood science capacity

Combating trade in illegal wood and forest products with machine learning

Trade in wood and forest products spans the global supply chain. Illegal logging and associated trade in forest products present a persistent threat to vulnerable ecosystems and communities. Illegal timber trade has been linked to violations of tax and conservation laws, as well as broader transnational crimes. The United States is the largest importer globally of wood and forest products, such as pulp, paper, flooring, and furniture-importing $78 billion in 2021. Transaction-level data such as shipping container manifests and bills of lading provide a comprehensive data source that can be used to detect and disrupt trade that may be suspected of containing illegally harvested or traded forest products. Owing to the volume, velocity, and complexity of shipment data, an automated decision support system is required for the purposes of detecting suspicious forest product shipments. We present a proof of concept framework using machine learning and big data approaches-combining domain expertise with automation-to achieve this objective. We formulated the underlying machine learning problem as an anomaly detection problem and collected and collated forest sector-specific domain knowledge to filter and target shipments of interest. In this work, we provide the overview of our framework, with the details of domain knowledge extraction and machine learning models, and discuss initial results and analysis of flagged anomalous and potentially suspicious records to demonstrate the efficacy of this approach. The proof of concept work presented here provides the groundwork for an actionable and feasible approach to assisting enforcement agencies with the detection of suspicious shipments that may contain illegally harvested or traded wood.

Timber DNA release using focused ultrasound extraction (FUSE) for genetic species identification

The use of genetic data for timber species and population assignment is a powerful tool for combating the illegal timber trade, but the challenges of extracting DNA from timber have prevented the routine use of genetics as a supply chain management tool. To overcome these challenges, we explored the feasibility of focused ultrasound extraction (FUSE) for rapid DNA release from timber. Using high-pressure ultrasound pulses, FUSE generates a cavitation bubble cloud that disintegrates samples into acellular debris, resulting in the mechanical release of DNA. In this work, FUSE was applied to white oak (Quercus alba) timber shavings to test the feasibility of using FUSE for timber DNA extraction for the first time. Results showed that FUSE processing disintegrated the tissue samples and released significant quantities of DNA. After five minutes of tissue processing DNA quantities of 0.21 ± 0.02 ng/mg, 0.99 ± 0.32 ng/mg, and 0.14 ± 0.01 ng/mg, were released from medium, coarse, and combination shaving groups, respectively. Amplification and sequencing of regions within the matK and rbcL chloroplast genes confirmed that the quality of DNA prepared with FUSE was suitable for PCR and short-read sequencing applications. Overall, these results show that FUSE can serve as a DNA sample preparation method capable of releasing high-quality DNA from timber in a fraction of the time required by conventional extraction methods. Based on the improved efficiency of DNA release with FUSE, ongoing work aims to develop this technology into portable systems that can be used to rapidly prepare timber samples for genetic species identification.

How to discriminate wood of CITES-listed tree species from their look-alikes: using an attention mechanism with the ResNet model on an enhanced macroscopic image dataset

Introduction

Global illegal trade in timbers is a major cause of the loss of tree species diversity. The Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) has been developed to combat the illegal international timber trade. Its implementation relies on accurate wood identification techniques for field screening. However, meeting the demand for timber field screening at the species level using the traditional wood identification method depending on wood anatomy is complicated, time-consuming, and challenging for enforcement officials who did not major in wood science.

Methods

This study constructed a CITES-28 macroscopic image dataset, including 9,437 original images of 279 xylarium wood specimens from 14 CITES-listed commonly traded tree species and 14 look-alike species. We evaluated a suitable wood image preprocessing method and developed a highly effective computer vision classification model, SE-ResNet, on the enhanced image dataset. The model incorporated attention mechanism modules [squeeze-and-excitation networks (SENet)] into a convolutional neural network (ResNet) to identify 28 wood species.

Results

The results showed that the SE-ResNet model achieved a remarkable 99.65% accuracy. Additionally, image cropping and rotation were proven effective image preprocessing methods for data enhancement. This study also conducted real-world identification using images of new specimens from the timber market to test the model and achieved 82.3% accuracy.

Conclusion

This study presents a convolutional neural network model coupled with the SENet module to discriminate CITES-listed species with their look-alikes and investigates a standard guideline for enhancing wood transverse image data, providing a practical computer vision method tool to protect endangered tree species and highlighting its substantial potential for CITES implementation.

Wood identification of Cyclobalanopsis (Endl.) Oerst based on microscopic features and CTGAN-enhanced explainable machine learning models

Introduction

Accurate and fast identification of wood at the species level is critical for protecting and conserving tree species resources. The current identification methods are inefficient, costly, and complex.

Methods

A wood species identification model based on wood anatomy and using the Cyclobalanopsis genus wood cell geometric dataset was proposed. The model was enhanced by the CTGAN deep learning algorithm and used a simulated cell geometric feature dataset. The machine learning models BPNN and SVM were trained respectively for recognition of three Cyclobalanopsis species with simulated vessel cells and simulated wood fiber cells.

Results

The SVM model and BPNN model achieved recognition accuracy of 96.4% and 99.6%, respectively, on the real dataset, using the CTGAN-generated vessel dataset. The BPNN model and SVM model achieved recognition accuracy of 75.5% and 77.9% on real dataset, respectively, using the CTGAN-generated wood fiber dataset.

Discussion

The machine learning model trained based on the enhanced cell geometric feature data by CTGAN achieved good recognition of Cyclobalanopsis, with the SVM model having a higher prediction accuracy than BPNN. The machine learning models were interpreted based on LIME to explore how they identify tree species based on wood cell geometric features. This proposed model can be used for efficient and cost-effective identification of wood species in industrial applications.

A DNA barcode reference library for CITES listed Malagasy Dalbergia species

On Madagascar, the illegal and unsustainable exploitation and illegal international trade of Dalbergia (rosewood) precious woods remain a serious conservation problem. Members of this genus are at high risk of extinction as a consequence of logging, mining, and slash and burn agriculture. Morphological identification of these Malagasy species is difficult in the absence of flowers and fruits, especially in the case of cut trees, sawn wood, and finished product. In this study, we use molecular barcoding to identify the Dalbergia species with the intent to contribute to the control of their illegal trade. Thirty-six Dalbergia samples representing 12 Malagasy species of which 11 have high commercial value, were collected to test the efficacy of a region of the plastid genome (rbcL) and a nuclear-transcribed ITS for barcoding. These widely used markers, as well as DNA barcoding gaps, "best match" and "best close match" approaches, and the neighbor-joining method were employed. All samples were amplified and sequenced using the two markers. Using a single locus, the "best match" and "best close match" approaches revealed that ITS has high discriminatory power within the tested Malagasy species. The combination of rbcL + ITS revealed 100% species discrimination. This study confirms that ITS alone and in combination with chloroplast barcode rbcL allow non-ambiguous identification for the 12 species studied. The results contribute to the development of DNA barcoding as a useful tool to identify Malagasy Dalbergia and suggest that the approach developed should be expanded to all 56 potentially exploited species in reference to international CITES requirements and the sustainable management of valuable resources.

Tropical Wood Species Recognition: A Dataset of Macroscopic Images

Forests are of incalculable value due to the ecosystem services they provide to humanity such as carbon storage, climate regulation and participation in the hydrological cycle. The threat to forests grows as the population increases and the activities that are carried out in it, such as: cattle rearing, illegal trafficking, deforestation and harvesting. Moreover, the environmental authorities do not have sufficient capacity to exercise strict control over wood production due to the vast variety of timber species within the countries, the lack of tools to verify timber species in the supply chain and the limited available and labelled digital data of the forest species. This paper presents a set of digital macroscopic images of eleven tropical forest species, which can be used as support at checkpoints, to carry out studies and research based on macroscopic analysis of cross-sectional images of tree species such as: dendrology, forestry, as well as algorithms of artificial intelligence. Images were acquired in wood warehouses with a digital magnifying glass following a protocol used by the Colombian Ministry of Environment, as well as the USA Forest Services and the International Association of Wood Anatomists. The dataset contains more than 8000 images with resolution of 640 × 480 pixels which includes 3.9 microns per pixel, and an area of (2.5 × 1.9) square millimeters where the anatomical features are exposed. The dataset presents great usability for academics and researchers in the forestry sector, wood anatomists and personnel who work with computational models, without neglecting forest surveillance institutions such as regional autonomous corporations and the Ministry of the Environment.

Study on the effect of income perception on cleaner-production fraud

We investigated a single-cycle product supply chain in a game model, where a supplier is the leader. By innovatively introducing cleaner production fraudulent income perception factor into a game model, we studied the mechanism of the effect of enterprise social responsibility and environmental awareness on cleaner production fraud. The results showed that the income of retailers and suppliers changes under different perceptions of fraudulent income. That is, the value of cleaner production fraudulent income perception factor will affect the enterprise's choice of differentiation strategy. When the enterprise's sense of social responsibility is weak, it is more likely to choose cleaner production fraud. Conversely, under the constraints of high social responsibility, it more likely avoids production fraud. Regarding government supervision, the income of suppliers and retailers changes under different government penalties. Furthermore, a reasonable punishment for cleaner production fraud can reduce such violations. However, after the punishment reaches a level, the efficiency of supervision begins to decline. In views of that, improving enterprise social responsibility through institutional reform is a more effective way to reduce cleaner production fraud. To contribute to a healthy competitive market environment, government supervision should establish a feedback mechanism, and make timely adjustments.

Organellar microcapture to extract nuclear and plastid DNA from recalcitrant wood specimens and trace evidence

BACKGROUND

Illegal logging is a global crisis with significant environmental, economic, and social consequences. Efforts to combat it call for forensic methods to determine species identity, provenance, and individual identification of wood specimens throughout the forest products supply chain. DNA-based methodologies are the only tools with the potential to answer all three questions and the only ones that can be calibrated "non-destructively" by using leaves or other plant tissue and take advantage of publicly available DNA sequence databases. Despite the potential that DNA-based methods represent for wood forensics, low DNA yield from wood remains a limiting factor because, when compared to other plant tissues, wood has few living DNA-containing cells at functional maturity, it often has PCR-inhibiting extractives, and industrial processing of wood degrades DNA. To overcome these limitations, we developed a technique-organellar microcapture-to mechanically isolate intact nuclei and plastids from wood for subsequent DNA extraction, amplification, and sequencing.

RESULTS

Here we demonstrate organellar microcapture wherein we remove individual nuclei from parenchyma cells in wood (fresh and aged) and leaves of Carya ovata and Tilia americana, amyloplasts from Carya wood, and chloroplasts from kale (Brassica sp.) leaf midribs. ITS (773 bp), ITS1 (350 bp), ITS2 (450 bp), and rbcL (620 bp) were amplified via polymerase chain reaction, sequenced, and heuristic searches against the NCBI database were used to confirm that recovered DNA corresponded to each taxon.

CONCLUSION

Organellar microcapture, while too labor-intensive for routine extraction of many specimens, successfully recovered intact nuclei from wood samples collected more than sixty-five years ago, plastids from fresh sapwood and leaves, and presents great potential for DNA extraction from recalcitrant plant samples such as tissues rich in secondary metabolites, old specimens (archaeological, herbarium, and xylarium specimens), or trace evidence previously considered too small for analysis.

Caveat emptor: On the Need for Baseline Quality Standards in Computer Vision Wood Identification

Computer vision wood identification (CVWID) has focused on laboratory studies reporting consistently high model accuracies with greatly varying input data quality, data hygiene, and wood identification expertise. Employing examples from published literature, we demonstrate that the highly optimistic model performance in prior works may be attributed to evaluating the wrong functionality—wood specimen identification rather than the desired wood species or genus identification—using limited datasets with data hygiene practices that violate the requirement of clear separation between training and evaluation data. Given the lack of a rigorous framework for a valid methodology and its objective evaluation, we present a set of minimal baseline quality standards for performing and reporting CVWID research and development that can enable valid, objective, and fair evaluation of current and future developments in this rapidly developing field. To elucidate the quality standards, we present a critical revisitation of a prior CVWID study of North American ring-porous woods and an exemplar study incorporating best practices on a new dataset covering the same set of woods. The proposed baseline quality standards can help translate models with high in silico performance to field-operational CVWID systems and allow stakeholders in research, industry, and government to make informed, evidence-based modality-agnostic decisions.

Honest nutraceuticals, cosmetics, therapies, and foods (NCTFs): standardization and safety of natural products

With the increasing demand for natural products by the consumer in the marketplace it is necessary to see a proportional increase in behind-the-scenes science to ensure that the ideology of safety and honesty, that is justifiably expected by the wider public, is adequately satisfied. It is of essence to have a fair yet firm governance of nutraceuticals, cosmetics, therapies, and foods. However, with increasing sophistications in adulteration and "claim" loopholes that make it easier for adulterated or counterfeited natural products to be "fudged" to meet the pharmacopeia standards, governance protocols must utilize an "identification and authentication" approach that goes beyond the Pharmacopeia standards to help regulate and transparently communicate natural products in the commercial context. While it is becoming a rat race in keeping commercial natural products honest, modern technology can support authenticators and adequately defeat these challenges.

Towards Sustainable North American Wood Product Value Chains, Part I: Computer Vision Identification of Diffuse Porous Hardwoods

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.

The XyloTron: Flexible, Open-Source, Image-Based Macroscopic Field Identification of Wood Products

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.

North American Hardwoods Identification Using Machine-Learning

This technical note determines the feasibility of using an InceptionV4_ResNetV2 convolutional neural network (CNN) to correctly identify hardwood species from macroscopic images. The method is composed of a commodity smartphone fitted with a 14× macro lens for photography. The end-grains of ten different North American hardwood species were photographed to create a dataset of 1869 images. The stratified 5-fold cross-validation machine-learning method was used, in which the number of testing samples varied from 341 to 342. Data augmentation was performed on-the-fly for each training set by rotating, zooming, and flipping images. It was found that the CNN could correctly identify hardwood species based on macroscopic images of its end-grain with an adjusted accuracy of 92.60%. With the current growing of machine-learning field, this model can then be readily deployed in a mobile application for field wood identification.

Machine Learning Models with Quantitative Wood Anatomy Data Can Discriminate between Swietenia macrophylla and Swietenia mahagoni

Illegal logging and associated trade aggravate the over-exploitation of Swietenia species, of which S. macrophylla King, S. mahagoni (L.) Jacq, and S. humilis Zucc. have been listed in Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) Appendix Ⅱ. Implementation of CITES necessitates the development of efficient forensic tools to identify wood species accurately, and ideally ones readily deployable in wood anatomy laboratories across the world. Herein, a method using quantitative wood anatomy data in combination with machine learning models to discriminate between three Swietenia species is presented, in addition to a second model focusing only on the two historically more important species S. mahagoni and S. macrophylla. The intra- and inter-specific variations in nine quantitative wood anatomical characters were measured and calculated based on 278 wood specimens, and four machine learning classifiers—Decision Tree C5.0, Naïve Bayes (NB), Support Vector Machine (SVM), and Artificial Neural Network (ANN)—were used to discriminate between the species. Among these species, S. macrophylla exhibited the largest intraspecific variation, and all three species showed at least partly overlapping values for all nine characters. SVM performed the best of all the classifiers, with an overall accuracy of 91.4% and a per-species correct identification rate of 66.7%, 95.0%, and 80.0% for S. humilis, S. macrophylla, and S. mahagoni, respectively. The two-species model discriminated between S. macrophylla and S. mahagoni with accuracies of over 90.0% using SVM. These accuracies are lower than perfect forensic certainty but nonetheless demonstrate that quantitative wood anatomy data in combination with machine learning models can be applied as an efficient tool to discriminate anatomically between similar species in the wood anatomy laboratory. It is probable that a range of previously anatomically inseparable species may become identifiable by incorporating in-depth analysis of quantitative characters and appropriate statistical classifiers.