Automated identification of ivory carvings using machine learning: A proof-of-concept approach to detecting the illegal wildlife trade

Andrius Tamosiunas; Simone Mozzon; David Bacon; Caroline Cox; Alan Collins

cs.CVOct 23, 2026

Automated identification of ivory carvings using machine learning: A proof-of-concept approach to detecting the illegal wildlife trade

Authors: Andrius Tamosiunas, Simone Mozzon, David Bacon, Caroline Cox, Alan Collins

Organizations: Institute of Theoretical Astrophysics, University of Oslo, PO Box 1029, Blindern 0315, Norway · CERCA/ISO, Department of Physics, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH, 44106, USA · Institute of Cosmology and Gravitation, University of Portsmouth, Dennis Sciama Building, Burnaby Road, Portsmouth, PO1 3FX, UK · School of Law, Faculty of Business and Law, University of Portsmouth, Richmond Building, Portland St, Portsmouth, PO1 3DE, UK ·

Abstract

Despite a ban on the international trading of recently poached ivory, elephants are still killed for their tusks. Ivory and its products are mainly sold in online marketplaces. However, due to the huge volume and variety of adverts and the similarity of ivory with other materials that can be legally sold, it is hard for enforcement officials to identify illicit trading and prosecute sellers. Existing techniques for identifying elephant ivory, such as Fourier-transform infrared spectroscopy and microscopy, can reliably distinguish ivory from other materials, but require physical access to the item and cannot be applied to photographs alone. As a result, these methods are unsuitable for screening the large volume of images posted in online marketplaces, leaving a gap for automated, image-based identification tools. In this proof of concept work, we investigate the potential of machine learning techniques for image analysis to automatically distinguish ivory objects from bone. We employ a semantic segmentation framework using a U-Net architecture with a ResNet34 backbone, leveraging transfer learning on an augmented dataset of 541 original images. Using a dataset of images of ivory and items resembling carved ivory, several models were trialled. On a distinct test set, the classifier achieved an Area Under the Curve (AUC) of 0.81 and an overall classification accuracy of 81% on confident samples. The paper presents the best-performing model as a proof of concept for developing a system that can effectively identify ivory and its products when indicative features of elephant ivory are present in an image. This work provides initial steps towards a machine learning-based framework designed to advise law-enforcement in detecting illegal elephant ivory trade.

Figures & tables

Panel of Figure 1 — Fig 1: Example of semantic segmentation. The original images are shown on the left, while the images on the right show the classification targets. Red pixels correspond to the ivory class, green pixels to non-ivory, and black pixels to the background (images from Wikimedia Commons under the public domain license).

Fig 2: The U-Net architecture used for image segmentation. The U-net architecture consists of a contracting path (left side) made of down-sampling units, followed by an expansive path (right side) consisting of up-sampling units. Each down-sampling unit consists of convolution layers that extract features from the input images, and a max-pooling layer to reduce the size of the feature map. The extracted features are passed down to the next down-sampling unit and the corresponding up-sampling unit.

Fig 3: Receiver operating characteristic (ROC) curve for the ivory classifier. The area under the ROC curve is 0.81. The model performs significantly better than a random classifier (shown by the dashed line).

Model	Precision	Recall	F1-Score	AUC	Youden’s J
Model I	0.46	0.88	0.60	0.75	0.43
Model II	0.64	0.73	0.68	0.77	0.52
Model III	0.68	0.75	0.72	0.81	0.55
Model IV	0.45	0.92	0.61	0.80	0.50

Table 1: Model performance metrics for ivory classification for different image augmentation strategies. The statistics here were calculated including all the images in the test dataset (those deemed certain and uncertain by the algorithm). See the Dataset section for a full description of the different models. The best performing model is marked in bold.

Model	Precision	Recall	F1-Score	AUC	Youden’s J
Model I	0.68	0.94	0.79	0.68	0.35
Model II	0.67	0.77	0.71	0.79	0.55
Model III	0.82	0.85	0.84	0.86	0.67
Model IV	0.57	0.98	0.72	0.78	0.49

Table 2: Model performance metrics for ivory classification for different image augmentation strategies. In this case we only consider test predictions that were marked as confident (i.e., for which the averaged ivory probability excess score was higher or equal to 0.3). The best-scoring model is marked in bold.

Fig 4: The confusion matrices for the different models (different image augmentation strategies) evaluated on the full test dataset.

Fig 5: The confusion matrices for the different models evaluated for confident predictions ( $|\hat{p}|>0.3$ ) only.

Fig 6: A selection of carved ivory images along with their pixel-level classification. The colors indicate the pixel classification (darker for bone, lighter for ivory and white for background). The figure legends also show the predicted class and the fraction of pixels classified as ivory.

Fig 7: A selection of bone images from our test dataset. The colorbar and legend conventions are the same as in Fig 6 .

Image Classification

Image classification assigns semantic labels to images, enabling recognition of objects, scenes, conditions, and fine-grained categories.

See all 386 papers

From Birdsong to Rumbles: Classifying Elephant Calls with Out-of-Species Embeddings
Apr 30, 2026Christiaan M. Geldenhuys, Thomas R. NieslerAudio Representation Learning Audio Understanding
Department of Electrical and Electronic Engineering University of Stellenbosch
Beyond Humans: Multispecies Animal Face Recognition Using Transfer Learning
Jun 8, 2026Maria De Marsico, Anil K. Jain, Annalaura MiglinoBiometric Identification Transfer Learning
Department of Computer Science, Sapienza University of Rome, Via Salaria, 113, Rome, 00185, State, Italy. · Department of Computer Science & Engineering, Michigan State University, Michigan State University Engineering Building 428 S. Shaw Lane, East Lansing, MI 48824, Michigan, US. · Department of Computer Science, University of Salerno, Via Giovanni Paolo II, 132, Fisciano (SA), 84084, State, Italy.
A Remote Approach to Cashew Orchard Detection: Leveraging Active Learning with Satellite Imagery in Guinea-Bissau
Aug 12, 2026Miguel Pereira, Sofia C. Pereira, Maria J. P. Vasconcelos +3Agricultural Remote Sensing

Apr 30, 2026J/K move · Enter open · S save

From Birdsong to Rumbles: Classifying Elephant Calls with Out-of-Species Embeddings

Christiaan M. Geldenhuys, Thomas R. Niesler

Department of Electrical and Electronic Engineering University of Stellenbosch

Audio Representation Learning Audio Understanding Bioacoustics Audio Classification

Open paper AskPDF arXiv

We show that pretrained acoustic embeddings classify elephant vocalisations at a level approaching that of end-to-end supervised neural networks, without any fine-tuning of the embedding model. This result is of practical importance because annotated bioacoustic data are scarce and costly to obtain, leaving conventional supervised approaches prone to overfitting and to poor generalisation under domain shift. A broad range of embedding models drawn from general audio, speech, and bioacoustic domains is evaluated, all of which are either out-of-domain (containing no bioacoustic data) or out-of-species (containing no elephant call data). The embedding networks themselves remain fixed; only the lightweight downstream classifiers, which include a linear model and several small neural networks, are trained. Among the models considered, Perch 2.0 achieves the best cross-validated classification performance, attaining AUCs of 0.849 on African bush elephant (Loxodonta africana) calls and 0.936 on Asian elephant (Elephas maximus) calls, with Perch 1.0 close behind. The best-performing system is within 2.2 % of an end-to-end supervised elephant call classification system. A layerwise analysis of pretrained transformer encoders, considered as embedding models, shows that intermediate representations outperform final-layer outputs. The second layer of both wav2vec2.0 and HuBERT encodes sufficient information for effective elephant call classification; truncation at this layer therefore preserves classification performance whilst retaining only approximately 10 % of the parameters of the full network. Such compact embedding networks are well suited to on-device processing where computational resources are limited.

Abstract

Image Classification

Explore similar work

From Birdsong to Rumbles: Classifying Elephant Calls with Out-of-Species Embeddings

Beyond Humans: Multispecies Animal Face Recognition Using Transfer Learning

A Remote Approach to Cashew Orchard Detection: Leveraging Active Learning with Satellite Imagery in Guinea-Bissau