Search

Menu
Search
in Ecology

Improving wildlife track classification through human-in-the-loop method and explainable AI

199 Views


Abstract

In this study, we present the integration of tracker expertise with artificial intelligence (AI) for wildlife species classification and its application to human-in-the-loop with an investigation in explainable AI. We collected images of wildlife tracks, built AI models from the track images, classified species based on the best performing model, and expert trackers evaluated the results against the AI model. The wildlife species included black rhinoceros (Diceros bicornis), blue wildebeest (Connochaetes taurinus), giraffe (Giraffa camelopardalis), and white rhinoceros (Ceratotherium simum). Two expert trackers and one non-expert tracker ranked the image quality of 3039 tracks. We then trained the AI model with different number of training images per class using different hyperparameter settings. The best-performing AI model was chosen and evaluated. Afterwards, 36 expert trackers evaluated the resulting model: one set using raw images only, and another set using raw with heatmap images. For the same hyperparameter settings with the best model performance evaluation, our method considerably increased the mean average precision@50–95 by 10.42% against a non-expert tracker. In addition, the required number of images for model training can be reduced by 25% when using inputs from a highly skilled expert tracker. The visual heatmaps provided a means in performing explainable AI by visually presenting the features of the tracks based on colours that help to guide the expert tracker evaluation.

Similar content being viewed by others

Iterative human and automated identification of wildlife images

Article

18 October 2021

Addressing significant challenges for animal detection in camera trap images: a novel deep learning-based approach

Article
Open access
09 May 2025

Man versus machine: cost and carbon emission savings of 4G-connected Artificial Intelligence technology for classifying species in camera trap images

Article
Open access
24 June 2024

Data availability

The wildlife track dataset analysed in the current study is not publicly available due to research permit condition number ENT 8/36/4 LVII (13) issued by the Botswana Ministry of Environment and Tourism. An ethical clearance certificate number HREC-12 was awarded by Botswana International University of Science and Technology. Sample wildlife species track classification and their heatmaps for the best performing AI model are shared at the links: https://github.com/Tinao76/Wildlife-Species-Track-Classification/tree/main/Sample%20AI%20Predictions and https://github.com/Tinao76/Wildlife-Species-Track-Classification/tree/main/Sample%20Heatmap%20Predictions respectively. The sample tracker assessment output of the 200 images for species track selection, species feature selections, hind laterality selections AI prediction evaluation and time duration are shared at https://github.com/Tinao76/Wildlife-Species-Track-Classification/tree/main/Post%20Tracker%20Assessment.

References

  1. Isabelle, D. A. & Westerlund, M. A review and categorization of artificial intelligence-based opportunities in wildlife, ocean and land conservation. Sustainability 14, 1979 (2022).

    Google Scholar 

  2. Corcoran, E., Winsen, M., Sudholz, A. & Hamilton, G. Automated detection of wildlife using drones: Synthesis, opportunities and constraints. Methods Ecol. Evol. 12, 1103–1114 (2021).

    Google Scholar 

  3. Schneider, S., Greenberg, S., Taylor, G. W. & Kremer, S. C. Three critical factors affecting automated image species recognition performance for camera traps. Ecol. Evol. 10, 3503–3517 (2020).

    Google Scholar 

  4. Raza, A., Hanif, F. & Mohammed, H. A. Analyzing the enhancement of CNN-YOLO and transformer based architectures for real-time animal detection in complex ecological environments. Sci. Rep. 15, 39142 (2025).

    Google Scholar 

  5. Petso, T., Jamisola, R. S. Jr., Mpoeleng, D., Bennitt, E. & Mmereki, W. Automatic animal identification from drone camera based on point pattern analysis of herd behaviour. Eco. Inform. 66, 101485 (2021).

    Google Scholar 

  6. Hua, A. et al. Protecting endangered megafauna through AI analysis of drone images in a low-connectivity setting: a case study from Namibia. PeerJ 10, e13779 (2022).

    Google Scholar 

  7. Duporge, I. et al. Baboonland dataset: Tracking primates in the wild and automating behaviour recognition from drone videos: I. duporge et al. Int. J. Comput. Vis. 133, 6578–6589 (2025).

  8. Duporge, I. et al. Automated rhinoceros detection in satellite imagery using deep learning. Sci. Rep. 15, 39352 (2025b).

    Google Scholar 

  9. Fergus, P., Chalmers, C., Longmore, S. & Wich, S. Harnessing artificial intelligence for wildlife conservation. Conservation 4, 685–702 (2024).

    Google Scholar 

  10. Shivaprakash, K. N. et al. Potential for artificial intelligence (AI) and machine learning (ML) applications in biodiversity conservation, managing forests, and related services in India. Sustainability 14, 7154 (2022).

    Google Scholar 

  11. Petso, T. & Jamisola, R. S. Jr. Wildlife conservation using drones and artificial intelligence in Africa. Sci. Robot. 8, eadm7008 (2023).

  12. Adadi, A. & Berrada, M. Peeking inside the black-box: A survey on explainable artificial intelligence (XAI). IEEE Access 6, 52138–52160 (2018).

    Google Scholar 

  13. Hassija, V. et al. Interpreting black-box models: A review on explainable artificial intelligence. Cogn. Comput. 16, 45–74 (2024).

    Google Scholar 

  14. Senoner, J., Schallmoser, S., Kratzwald, B., Feuerriegel, S. & Netland, T. Explainable AI improves task performance in human-AI collaboration. Sci. Rep. 14, 31150 (2024).

    Google Scholar 

  15. Kawai, Y. et al. Explainable artificial intelligence-based prediction of poor neurological outcome from head computed tomography in the immediate post-resuscitation phase. Sci. Rep. 13, 5759 (2023).

    Google Scholar 

  16. Aldoseri, A., Al-Khalifa, K. N. & Hamouda, A. M. Re-thinking data strategy and integration for artificial intelligence: concepts, opportunities, and challenges. Appl. Sci. 13, 7082 (2023).

    Google Scholar 

  17. Fortson, L., Crowston, K., Kloetzer, L. & Ponti, M. Artificial intelligence and the future of citizen science. Citizen Sci. Theory Pract. 9 (2024).

  18. Pardo, L. E. et al. Snapshot Safari: A large-scale collaborative to monitor Africa’s remarkable biodiversity. S. Afr. J. Sci. 117, 1–4 (2021).

    Google Scholar 

  19. Kumar, S. et al. Applications, challenges, and future directions of human-in-the-loop learning. IEEE Access 12, 75735–75760 (2024).

    Google Scholar 

  20. Ray, S. & Ray, R. L. Incorporating indigenous knowledge systems into AI governance: Enhancing ethical frameworks with Maori and Navajo perspectives. AI Ethics 6, 114 (2026).

    Google Scholar 

  21. Bohnett, E. et al. Human expertise combined with artificial intelligence improves performance of snow leopard camera trap studies. Glob. Ecol. Conserv. 41, e02350 (2023).

    Google Scholar 

  22. Kapuka, A. & Hlásny, T. Climate change impacts on ecosystems and adaptation options in nine countries in southern Africa: What do we know?. Ecosphere 12, e03860 (2021).

    Google Scholar 

  23. Stoldt, M., Göttert, T., Mann, C. & Zeller, U. Transfrontier conservation areas and human-wildlife conflict: The case of the Namibian component of the Kavango-Zambezi (KAZA) TFCA. Sci. Rep. 10, 7964 (2020).

    Google Scholar 

  24. Feldmeier, D. E., Schmitz, O. J., Carter, N. H., Masunga, G. S. & Orrick, K. D. Predictions & perceptions: A social-ecological analysis of human-carnivore conflict in Botswana. Biol. Conserv. 294, 110615 (2024).

    Google Scholar 

  25. Gulati, S., Karanth, K. K., Le, N. A. & Noack, F. Human casualties are the dominant cost of human-wildlife conflict in India. Proc. Natl. Acad. Sci. 118, e1921338118 (2021).

    Google Scholar 

  26. Pozo, R. A. et al. A multispecies assessment of wildlife impacts on local community livelihoods. Conserv. Biol. 35, 297–306 (2021).

    Google Scholar 

  27. Sianga, K. et al. Patterns of spatial dynamics and distribution of African elephants (Loxodonta africana) in the Central Kalahari Game Reserve, Botswana. Glob. Ecol. Conserv. 56, e03284 (2024).

    Google Scholar 

  28. Petso, T., Jamisola, R. S. Jr. & Mpoeleng, D. Review on methods used for wildlife species and individual identification. Eur. J. Wildl. Res. 68, 3 (2022).

    Google Scholar 

  29. Kistner, F., Slaney, L., Jewell, Z., Ben-David, A. & Alibhai, S. It’s otterly confusing-distinguishing between footprints of three of the four sympatric Asian otter species using morphometrics and machine learning. OTTER J. Int. Otter Surviv. Fund 2022, 108–125 (2022).

    Google Scholar 

  30. Kistner, F., Slaney, L. & Morant, N. Can you tell the species by a footprint?-identifying three of the four sympatric southeast Asian otter species using computer vision and deep learning. IUCN Otter Spec. Group Bull 40, 197–210 (2023).

    Google Scholar 

  31. Alibhai, S. K. et al. ‘I know the tiger by his paw’: A non-invasive footprint identification technique for monitoring individual Amur tigers (panthera tigris altaica) in snow. Eco. Inform. 73, 101947 (2023).

    Google Scholar 

  32. Alibhai, S. K., Jewell, Z. C. & Law, P. R. A footprint technique to identify white rhino ceratotherium simum at individual and species levels. Endanger. Species Res. 4, 205–218 (2008).

    Google Scholar 

  33. Li, B. V., Alibhai, S., Jewell, Z., Li, D. & Zhang, H. Using footprints to identify and sex giant pandas. Biol. Conserv. 218, 83–90 (2018).

    Google Scholar 

  34. Shinoda, R. & Shiohara, K. Openanimaltracks: A dataset for animal track recognition. In 2024 IEEE International Conference on Image Processing (ICIP), 110–116 (IEEE, 2024).

  35. Jewell, Z. C., Alibhai, S., Law, P. R., Uiseb, K. & Lee, S. Monitoring rhinoceroses in Namibia’s private custodianship properties. PeerJ 8, e9670 (2020).

    Google Scholar 

  36. Van den Heever, A., Mhlongo, R., Thomas, I. et al. Tracker Manual: A Practical Guide to Animal Tracking in Southern Africa (Penguin Random House South Africa, 2024).

  37. Liebenberg, L. The Art of Tracking the Origin of Science (Claremont, South Africa, David Philip, 1990).

    Google Scholar 

  38. Jewell, Z. C., Alibhai, S. K. & Law, P. R. Censusing and monitoring black rhino (Diceros bicornis) using an objective spoor (footprint) identification technique. J. Zool. 254, 1–16 (2001).

    Google Scholar 

  39. Alibhai, S., Jewell, Z. & Evans, J. The challenge of monitoring elusive large carnivores: An accurate and cost-effective tool to identify and sex pumas (Puma concolor) from footprints. PLoS One 12, e0172065 (2017).

    Google Scholar 

  40. Petso, T. et al. An investigation of lightweight AI models to classify African ungulate species from tracks. Ecol. Inform. 103393 (2025).

  41. Jewell, Z. C. et al. Spotting cheetahs: Identifying individuals by their footprints. J. Vis. Exp. JoVE 54034 (2016).

  42. Jessen, T. D., Ban, N. C., Claxton, N. X. & Darimont, C. T. Contributions of Indigenous Knowledge to ecological and evolutionary understanding. Front. Ecol. Environ. 20, 93–101 (2022).

    Google Scholar 

  43. Molnár, Z. & Babai, D. Inviting ecologists to delve deeper into traditional ecological knowledge. Trends Ecol. Evol. 36, 679–690 (2021).

    Google Scholar 

  44. Khosravi Mashizi, A. & Escobedo, F. The role of traditional ecological knowledge and ecosystem quality in managing ecosystem services. Sci. Rep. 15, 31510 (2025).

    Google Scholar 

  45. Fa, J. E. & Luiselli, L. Weaving the middle spaces between indigenous and scientific knowledge for biodiversity conservation and ecology (2025).

  46. Hsing, P.-Y. et al. Large-scale mammal monitoring: The potential of a citizen science camera-trapping project in the United Kingdom. Ecol. Solut. Evid. 3, e12180 (2022).

    Google Scholar 

  47. Green, S. E., Rees, J. P., Stephens, P. A., Hill, R. A. & Giordano, A. J. Innovations in camera trapping technology and approaches: The integration of citizen science and artificial intelligence. Animals 10, 132 (2020).

    Google Scholar 

  48. Law, P. R., Jewell, Z. & Alibhai, S. Using shape and size to quantify variation in footprints for individual identification: Case study with white rhinoceros (Ceratotherium simum). Wildl. Soc. Bull. 37, 433–438 (2013).

    Google Scholar 

  49. Tucker, J. M. et al. Development of a non-invasive method for species and sex identification of rare forest carnivores using footprint identification technology. Eco. Inform. 79, 102431 (2024).

    Google Scholar 

  50. Adablanu, S., Barman, U., Das, D. & Dweh, T. J. Transforming skin cancer detection with AI-based convolutional and transformer models. Iradiology 4, 51–62 (2026).

    Google Scholar 

  51. Fan, S. Explaining YOLOv9 object detection with attribution-based heatmap methods. In 2024 IEEE 2nd International Conference on Electrical, Automation and Computer Engineering (ICEACE), 201–206 (IEEE, 2024).

  52. Bhati, D., Neha, F. & Amiruzzaman, M. A survey on explainable artificial intelligence (XAI) techniques for visualizing deep learning models in medical imaging. J. Imaging 10, 239 (2024).

    Google Scholar 

  53. El-Sofany, H., Bouallegue, B. & El-Latif, Y. M. A. A proposed technique for predicting heart disease using machine learning algorithms and an explainable AI method. Sci. Rep. 14, 23277 (2024).

    Google Scholar 

  54. Dusza, P. et al. Comparative evaluation of CAM methods for enhancing explainability in veterinary radiography. Sci. Rep. 15, 29690 (2025).

    Google Scholar 

  55. Singh, A. K., Chaudhuri, D., Singh, M. P. & Chattopadhyay, S. Integrative CAM: Adaptive layer fusion for comprehensive interpretation of CNNs. arXiv preprint arXiv:2412.01354 (2024).

  56. Hammad, M. et al. Explainable AI for lung cancer detection via a custom CNN on CT images. Sci. Rep. 15, 12707 (2025).

    Google Scholar 

  57. Huebner, S. E., Palmer, M. S. & Packer, C. Human supervision is key to achieving accurate AI-assisted wildlife identifications in camera trap images. Citizen Sci. Theory Pract. 9 (2024).

  58. Chien, J.-C., Lee, J.-D., Hu, C.-S. & Wu, C.-T. The usefulness of gradient-weighted CAM in assisting medical diagnoses. Appl. Sci. 12, 7748 (2022).

    Google Scholar 

  59. Tang, D. et al. Reviewing CAM-based deep explainable methods in healthcare. Appl. Sci. 14, 4124 (2024).

    Google Scholar 

Download references

Acknowledgements

We thank the Botswana Defence Force as a collaborative institution that provided the research vehicle with an experienced off-road driver and military security escorts for the research team when in the field. We would sincerely like to thank the expert trackers for their great contribution namely, Adder Mate, Arnorld Kitso Modisane, Babui Mogotsi, Boikanyo Motlhatlhobi, Boikobo Fane, Disho Shakoii, Ditiro Scotch, David Sekudube, Emmanuel Radithipa, Ernest Makgoba, France Kamxee, Frenando Moronga, George David, Gofaone Matoteng, Gotlhophamang Sedibo, Itireleng Keletshabe, Ithuteng Keabereka, Jane Jackson, Jane Samasako, Kabo Banaeng, Kasu Samson Molaodi, Keipegetse Bombom, Kenosi Mbambo, Kesupamang Haita, Kethamile Million Orebonye, Lesego Malikopo, Letsomo Kauma, Mojaboswa Chisanga, Mosimanegape Kaelo, Mosweu Mosweu, Olebogeng Moalosi, Onkabetse Marumo, Omongwe Tebalo, Peter Moza Maburu, Philimon Ngwenyama, Pitso Kampa, Romeo Lewis, Ronnie Lewetse, Taolo Monchusi, Tefo Sekite, Thapelo Ramontshonyane, Thomas Mokoya, Tshepo Amos, Shepherd Samboko, Situmbeko Motsamai, Veivanga Kamutati. In addition, the designated drivers; Kago Petorose, Olebile Makgoloke, Omphemetse Motlhabaphuti and Tapiwa Baraedi. The authors would also like to express their gratitude to Khama Rhino Sanctuary and Mokolodi Nature Reserve Management for their support in the lodging facilities.

Funding

Research was sponsored by the United States Army Research Office and was accomplished under Grant Number W911NF-23-1-0293. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the U.S. Army Research Office or the U.S. Government. The U.S. Government is authorized to reproduce and distribute reprints for Government purposes notwithstanding any copyright notation herein.

Author information

Authors and Affiliations

  1. Department of Mechanical, Energy and Industrial Engineering, Botswana International University of Science and Technology, Private Bag 16, Palapye, Botswana

    Tinao Petso, Rodrigo S. Jamisola Jr., Molaletsa Namoshe & Wazha Mmereki

  2. WildTrack Inc., Durham, NC, USA

    Sky Alibhai & Zoe Jewell

  3. Duke University, Nicholas School of the Environment, Durham, NC, USA

    Sky Alibhai & Zoe Jewell

  4. Khama Rhino Sanctuary, P O Box 10, Serowe, Botswana

    Montiredi Lebopo

  5. Botswana Defence Force, Private Bag X06, Gaborone, Botswana

    Ketshabile Peter

Authors

  1. Tinao Petso
    View author publications

    Search author on:PubMed Google Scholar

  2. Rodrigo S. Jamisola Jr.
    View author publications

    Search author on:PubMed Google Scholar

  3. Sky Alibhai
    View author publications

    Search author on:PubMed Google Scholar

  4. Montiredi Lebopo
    View author publications

    Search author on:PubMed Google Scholar

  5. Ketshabile Peter
    View author publications

    Search author on:PubMed Google Scholar

  6. Molaletsa Namoshe
    View author publications

    Search author on:PubMed Google Scholar

  7. Wazha Mmereki
    View author publications

    Search author on:PubMed Google Scholar

  8. Zoe Jewell
    View author publications

    Search author on:PubMed Google Scholar

Contributions

Conceptualization – T.P, R.S.J.J, S.A, and Z.J; methodology – T.P, R.S.J.J, S.A, and Z.J; validation – T.P, R.S.J.J, S.A, M.L, K.P, and Z.J; investigation – T.P, R.S.J.J, S.A, M.L, K.P, and Z.J; data curation – T.P, R.S.J.J, S.A, M.L, K.P, and Z.J; formal analysis – T.P, R.S.J.J, S.A, and Z.J; writing: original draft preparation – T.P, R.S.J.J, S.A, and Z.J; writing: review and editing –T.P, R.S.J.J, S.A, N.M, W.M and Z.J; project supervision – R.S.J.J, S.A, N.M and Z.J; project administration – R.S.J.J; funding acquisition – R.S.J.J, S.A, and Z.J.

Corresponding author

Correspondence to
Tinao Petso.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Supplementary Information. (download PDF )

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Petso, T., Jamisola, R.S., Alibhai, S. et al. Improving wildlife track classification through human-in-the-loop method and explainable AI.
Sci Rep (2026). https://doi.org/10.1038/s41598-026-48229-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1038/s41598-026-48229-4

Keywords

  • Human-in-the-loop
  • Artificial intelligence models
  • Explainable AI
  • Species classification
  • Wildlife tracks

Source: Ecology - nature.com

Developmental and geographic transcriptomic variation in Anisakis simplex (s. s.) reveals lncRNA-mediated regulation of mRNA expression

Alkalinity-enhanced artificial substrates modulate local pH and increase survivorship of early-stage coral recruits

This portal is not a newspaper as it is updated without periodicity. It cannot be considered an editorial product pursuant to law n. 62 of 7.03.2001. The author of the portal is not responsible for the content of comments to posts, the content of the linked sites. Some texts or images included in this portal are taken from the internet and, therefore, considered to be in the public domain; if their publication is violated, the copyright will be promptly communicated via e-mail. They will be immediately removed.

Back to Top
Close