Routinely censusing rhinoceros’ populations is central to their conservation and protection from illegal killing. In Namibia, both white (Ceratotherium simum) and black (Diceros bicornis) rhinoceros occur on private land, in the latter case under a custodianship program of the Namibian Ministry of Environment and Tourism (MET). Black rhinoceros custodian landowners are responsible for the protection of the rhinoceroses on their land and are required to report regularly to the MET. Monitoring imposes a financial burden on custodians yet many of the techniques used involve expensive monitoring techniques that include the need for aerial support and/or animal instrumentation. During May and June 2018, WildTrack undertook a pilot study to census black and white rhinoceros on three private custodianship properties in Namibia. We tested three footprint identification methods for obtaining estimates of rhinoceros populations in an effort to provide less costly alternative monitoring options to rhinoceros custodians. The first was a full monitoring protocol with two components: (a) tracking each individual animal and matching them to their footprints, (b) identifying those individuals from the heel lines on the prints. The second method used simple visual heel line identification ex-situ, and the third method used just an objective footprint identification technique. These methods offer different options of fieldwork labour and cost and were designed to offer monitoring options to custodians that provided information about rhinoceros movement and location, with minimal disturbance to the rhinoceros, and best matched their human and economic resources. In this study, we describe the three methods and report the results of the pilot study to compare and evaluate their utility for rhinoceros monitoring. The first method successfully matched each trail photographed to a known rhinoceros at each site. When the other two methods disagreed with the first, they did so by failing to match single trails to a known rhinoceros, thereby creating fictitious identities consisting of a single trail. This failure occurred twice in one application, but otherwise at most once. We expect this failure can be eliminated through more stringent criteria for collecting photographs of footprints. We also briefly compare the use of footprint monitoring with other commonly used monitoring techniques. On this basis, landowners hosting rhinoceros can evaluate which method best suits their needs and resources.

Background/Objectives: This study investigated the use of artificial intelligence (AI) to identify and match barefoot prints belonging to the same individual on soft and sandy soil substrates. Recognizing footprints on soil is challenging due to low contrast and variability in impressions. Methods: We introduce Deep Learning Footprint Identification Technology (DeepFIT), based on a modified You Only Look Once (YOLOv11s) algorithm, using three methods, namely, Bounding Box (BBox), 16 anatomical landmarks, and automatically segmented outlines (Auto-Seg). An Extra Small Detection Head (XSDH) was added to improve feature extraction at smaller scales and enhance generalization through multi-scale supervision, reducing overfitting to specific spatial patterns. Results: Forty adults (20 males, 20 females) participated, with 600 images per individual. As the number of individuals in model training increased, the BBox model’s accuracy declined, resulting in misclassification on the test set. The average performance accuracy across both substrates was 77% for BBox, 90% for segmented outlines, and 96% for anatomical landmarks. Conclusions: The landmark method was the most reliable for identifying and matching barefoot prints on both soft and sandy soils. This approach can assist forensic practitioners in linking suspects to crime scenes and reconstructing events from footprint evidence, providing a valuable tool for forensic investigations.

This paper investigates human–AI collaboration through explainable AI where we interpret the results of barefoot print clustering using unsupervised machine learning. This can be used to identify the number of individuals from barefoot prints on the ground as a tool in forensics or anti-poaching. A self-supervised vision transformer, DINOv2, is used to automatically extract feature embeddings from localized barefoot-print regions to identify trails belonging to an individual on soil substrate. Furthermore, we introduce an Embedding Spatial Attribution Module (ESAM) to generate spatial attribution heatmaps, enabling visualization of discriminative regions that contribute to individual-specific trail identification and improving model explainability. The proposed method is named XTrail-ID, an explainable human footprint trail identification framework with two variants, OBB-XTrail-ID (oriented bounding box-based), and SEG-XTrail-ID (segmentation-based). We quantify embedding similarity using three complementary metrics: cosine similarity, Pearson correlation coefficient, and Spearman rank correlation. Twenty adults (ten males, ten females) participated, with a total of 1000 trail images extracted from UAV imagery. SEG-XTrail-ID using cosine similarity yielded the highest performance, with (3.21) discriminability and (94.2%) accuracy, while OBB-XTrail-ID using cosine similarity achieved (2.54) discriminability and (91.5%) accuracy. In addition, the latter exhibited reduced consistency in footprint grouping when more than three individuals were present within a single frame.