Machine learning to classify animal species in camera trap images: Applications in ecology
Open accessPublished version of a preprint
Abstract
Abstract 1. Motion-activated cameras (“camera traps”) are increasingly used in ecological and management studies for remotely observing wildlife and have been regarded as among the most powerful tools for wildlife research. However, studies involving camera traps result in millions of images that need to be analyzed, typically by visually observing each image, in order to extract data that can be used in ecological analyses. 2. We trained machine learning models using convolutional neural networks with the ResNet-18 architecture and 3,367,383 images to automatically classify wildlife species from camera trap images obtained from five states across the United States. We tested our model on an independent subset of images not seen during training from the United States and on an out-of-sample (or “out-of-distribution” in the machine learning literature) dataset of ungulate images from Canada. We also tested the ability of our model to distinguish empty images from those with animals in another out-of-sample dataset from Tanzania, containing a faunal community that was novel to the model. 3. The trained model classified approximately 2,000 images per minute on a laptop computer with 16 gigabytes of RAM. The trained model achieved 98% accuracy at identifying species in the United States, the highest accuracy of such a model to date. Out-of-sample validation from Canada achieved 82% accuracy, and correctly identified 94% of images containing an animal in the dataset from Tanzania. We provide an R package (Machine Learning for Wildlife Image Classification; MLWIC) that allows the users to A) implement the trained model presented here and B) train their own model using classified images of wildlife from their studies. 4. The use of machine learning to rapidly and accurately classify wildlife in camera trap images can facilitate non-invasive sampling designs in ecological studies by reducing the burden of manually analyzing images. We present an R package making these methods accessible to ecologists. We discuss the implications of this technology for ecology and considerations that should be addressed in future implementations of these methods.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The performance of a deep learning neural network improves logarithmically with increasing sample size.
“It is generally accepted that the performance of a deep learning neural network improves logarithmically with increasing sample size [74-77].”
From Automated Defect Detection and Decision-Support in Gas Turbine Blade Inspection (2021) · cites this paper as [75]
The Results section reports that species and groups with more training images were classified more accurately. Evaluating model accuracy says the authors modeled accuracy against the base-10 logarithm of training-image count, but fitting a generalized additive model to that predictor does not establish that performance itself improves logarithmically, or that this relationship holds for deep learning networks generally. No page number is available in our copy. (LLM, conf=0.86)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 74,248 · see citing works
Learning representations by back-propagating errors
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986) · Nature · doi:10.1038/323533a0 · Cited by 31,892 · see citing works
TensorFlow: A system for large-scale machine learningPreprint
Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M., Ghemawat, S., Irving, G., Isard, M., et al. (2016) · arXiv (Cornell University) · doi:10.5555/3026877.3026899 · Cited by 16,369 · see citing works
Advances in Neural Information Processing Systems 25
Zhang, Y., Sutton, C. A., Storkey, A., & Zoubin Ghahramani, A. (2012) · Cited by 12,653 · see citing works
Machine learning: Trends, perspectives, and prospects
Jordan, M. I., & Mitchell, T. M. (2015) · Science · doi:10.1126/science.aaa8415 · Cited by 9,955 · see citing works
The art of modelling range-shifting species: <i>The art of modelling range-shifting species</i>
Elith, J., Kearney, M., & Phillips, S. (2010) · Methods in Ecology and Evolution · doi:10.1111/j.2041-210x.2010.00036.x · Cited by 2,813 · see citing works
Geographical ecology; patterns in the distribution of species
MacArthur, R. H. (1972) · Medical Entomology and Zoology · Cited by 2,003 · see citing works
Norouzzadeh, M. S., Nguyen, A., Kosmala, M., Swanson, A., Palmer, M. S., Packer, C., & Clune, J. (2018) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1719367115 · Cited by 1,230 · see citing works
Self-citationCamera Traps in Animal Ecology
O’Connell, A. F., Karanth, K. U., & Nichols, J. D. (2010) · Springer eBooks · doi:10.1007/978-4-431-99495-4 · Cited by 768 · see citing works
Blumstein, D. T., Mennill, D. J., Clemins, P. J., Girod, L., Yao, K., Patricelli, G. L., Deppe, J. L., Krakauer, A. H., Clark, C. W., Cortopassi, K. A., et al. (2011) · Journal of Applied Ecology · doi:10.1111/j.1365-2664.2011.01993.x · Cited by 657 · see citing works
camtrapR: an R package for efficient camera trap data management
Niedballa, J., Sollmann, R., Courtiol, A., & Wilting, A. (2016) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.12600 · Cited by 551 · see citing works
Swanson, A., Kosmala, M., Lintott, C., Simpson, R., Smith, A., & Packer, C. (2015) · Scientific Data · doi:10.1038/sdata.2015.26 · Cited by 510 · see citing works
Observations in ecology: you can’t make progress on processes without understanding the patterns
Underwood, A. J., Chapman, M. G., & Connell, S. D. (2000) · Journal of Experimental Marine Biology and Ecology · doi:10.1016/s0022-0981(00)00181-7 · Cited by 461 · see citing works
Scaling‐up camera traps: monitoring the planet's biodiversity with networks of remote sensors
Steenweg, R. W., Hebblewhite, M., Kays, R., Ahumada, J., Fisher, J. T., Burton, A. C., Townsend, S. E., Carbone, C., Rowcliffe, J. M., Whittington, J., et al. (2016) · Frontiers in Ecology and the Environment · doi:10.1002/fee.1448 · Cited by 447 · see citing works
GENERALIZED SITE OCCUPANCY MODELS ALLOWING FOR FALSE POSITIVE AND FALSE NEGATIVE ERRORS
Royle, J. A., & Link, W. A. (2006) · Ecology · doi:10.1890/0012-9658(2006)87[835:gsomaf]2.0.co;2 · Cited by 402 · see citing works
Rovero, F., Zimmermann, F., Berzi, D., & Meek, P. D. (2013) · Bollettino del CILEA (CILEA) · doi:10.4404/hystrix-24.2-6316 · Cited by 376 · see citing works
Hierarchical Modeling and Inference in Ecology
(2009) · Elsevier eBooks · doi:10.1016/b978-0-12-374097-7.x0001-4 · Cited by 364 · see citing works
Gomez-Villa, A., Salazar, A., & Vargas‐Bonilla, J. F. (2017) · Ecological Informatics · doi:10.1016/j.ecoinf.2017.07.004 · Cited by 350 · see citing works
Distance sampling with camera traps
Howe, E. J., Buckland, S. T., Després‐Einspenner, M., & Kühl, H. S. (2017) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.12790 · Cited by 284 · see citing works
Swanson, A., Kosmala, M., Lintott, C., & Packer, C. (2016) · Conservation Biology · doi:10.1111/cobi.12695 · Cited by 251 · see citing works
Automated identification of animal species in camera trap images
Yu, X., Wang, J., Kays, R., Jansen, P. A., Wang, T., & Huang, T. S. (2013) · EURASIP Journal on Image and Video Processing · doi:10.1186/1687-5281-2013-52 · Cited by 232 · see citing works
Kelly, M. J., Noss, A. J., Bitetti, M. S. D., et al. (2008) · Journal of Mammalogy · doi:10.1644/06-mamm-a-424r.1 · Cited by 226 · see citing works
Tikhonov, G. N., Abrego, N., Dunson, D. B., et al. (2017) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.12723 · Cited by 220 · see citing works
Spatiotemporal hierarchical modelling of species richness and occupancy using camera trap data
Tobler, M. W., Hartley, A. C. Z., Carrillo‐Percastegui, S. E., & Powell, G. V. N. (2015) · Journal of Applied Ecology · doi:10.1111/1365-2664.12399 · Cited by 199 · see citing works
Deep convolutional neural network based species recognition for wild animal monitoring
Chen, G., Han, T. X., He, Z., Kays, R., & Forrester, T. D. (2014) · Proceedings - International Conference on Image Processing · doi:10.1109/icip.2014.7025172 · Cited by 190 · see citing works
Palmer, M. S., Fieberg, J., Swanson, A., Kosmala, M., & Packer, C. (2017) · Ecology Letters · doi:10.1111/ele.12832 · Cited by 172 · see citing works
A Novel Method to Reduce Time Investment When Processing Videos from Camera Trap Studies
Swinnen, K. R., Reijniers, J., Breno, M., et al. (2014) · PLoS ONE · doi:10.1371/journal.pone.0098881 · Cited by 89 · see citing works
Anthropogenic factors predict movement of an invasive species
Tabak, M. A., Piaggio, A. J., Miller, R. S., et al. (2017) · Ecosphere · doi:10.1002/ecs2.1844 · Cited by 84 · see citing works
Self-citationElias, A. R., Golubovic, N., Krintz, C., & Wolski, R. (2017) · IEEE/ACM International Conference on Internet-of-Things Design and Implementation (IoTDI) · doi:10.1145/3054977.3054986 · Cited by 80 · see citing works
On the Reliability of Expert Identification of Small-Medium Sized Mammals from Camera Trap Photos
Meek, P. D., Vernes, K., & Falzon, G. (2013) · Wildlife Biology in Practice · doi:10.2461/wbp.2013.9.4 · Cited by 60 · see citing works
Hanya, G., Otani, Y., Hongo, S., Honda, T., Okamura, H., & Higo, Y. (2018) · PLoS ONE · doi:10.1371/journal.pone.0190631 · Cited by 50 · see citing works
Interactions Between Carnivores in Madagascar and the Risk of Disease Transmission
Rasambainarivo, F., Farris, Z. J., Andrianalizah, H., et al. (2017) · EcoHealth · doi:10.1007/s10393-017-1280-7 · Cited by 33 · see citing works
Scott, A. B., Phalen, D. N., Hernández‐Jover, M., et al. (2018) · Avian Diseases · doi:10.1637/11761-101917-reg.1 · Cited by 29 · see citing works
Flagged first, then by citations
Research works printed but not linked: 12
- Brown, J. H. (1995). Macroecology. University of Chicago Press.
- Getzin, S., Wiegand, K., & Schöning, I. (2012). Assessing biodiversity in forests using very high-resolution images and unmanned aerial vehicles. Methods in Ecology and Evolution, 3(2), 397-404. doi:10.1111/j.2041-210X.2011.00158.x · DOI
- Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning (1st ed.). Cambridge, Massachusetts: MIT Press.
- certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
- Guillera-Arroita, G., Lahoz-Monfort, J. J., van Rooyen, A. R., Weeks, A. R., & Tingley, R. (2017). Dealing with false-positive and false-negative errors about species occurrence at multiple levels. Methods in Ecology and Evolution, 8(9), 1081-1091. doi:10.1111/2041- 210X.12743 · DOI
- Gurney, K. (2014). An Introduction to Neural Networks (1st ed.). London: CRC Press. Retrieved from https://www.taylorfrancis.com/books/9781482286991
- Lazebnik, S., Schmid, C., & Ponce, J. (2006). Beyond bags of features: spatial pyramid matching for recognizing natural scene categories. In Computer vision and pattern recognition (Vol. 2, pp. 2169-2178). New York: IEEE.
- MacKenzie, D. I., Nichols, J. D., Royle, J. A., Pollock, K. H., Bailey, L. L., & Hines, J. E. (2017). Occupancy Estimation and Modeling: Inferring Patterns and Dynamics of Species Occurrence (2nd ed.). London, UK: Academic Press.
- Mohri, M., Rostamizadeh, A., & Talwalkar, A. (2012). Foundations of Machine Learning. MIT Press.
- certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not this version posted July 9, 2018. ; https://doi.org/10.1101/346809 doi: bioRxiv preprint · DOI
- Tobler, M. W., Carrillo-Percastegui, S. E., Leite Pitman, R., Mares, R., & Powell, G. (2008). An evaluation of camera traps for inventorying large-and medium-sized terrestrial rainforest mammals. Animal Conservation, 11(3), 169-178. doi:10.1111/j.1469-1795.2008.00169.x certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not this version posted July 9, 2018. ; https://doi.org/10.1101/346809 doi: bioRxiv preprint · DOI
- certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.
Cited by
Sorted by citations
Applications for deep learning in ecology
Christin, S., Hervet, É., & Lecomte, N. (2019) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13256 · Cited by 724 · see citing works
WILDS: A Benchmark of in-the-Wild Distribution Shifts
Koh, P. W., Sagawa, S., Marklund, H., Xie, S. M., Zhang, M., Balsubramani, A., Hu, W., Yasunaga, M., Phillips, R. L., Gao, I., et al. (2020) · arXiv · doi:10.48550/arxiv.2012.07421 · Cited by 518 · see citing works
“…Parham et al. (2017), i.e., the number of distinct individual animals of each species seen in the overall collection of images; a task that is less realistic but still relevant and useful for ecologists might be to classify what species of animal is seen in each image Tabak et al. (2019).”
Shahinfar, S., Meek, P., & Falzon, G. (2020) · Ecological Informatics · doi:10.1016/j.ecoinf.2020.101085 · Cited by 229 · see citing works
Deep learning‐based methods for individual recognition in small birds
Ferreira, A. C., Silva, L. R., Renna, F., Brandl, H. B., Renoult, J. P., Farine, D. R., Covas, R., & Doutrelant, C. (2020) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13436 · Cited by 203 · see citing works
“This and other examples (e.g. Rzanny, Seeland, Wäldchen & Mäder, 2017; Tabak et al., 2019) highlight the potentialities of deep learning for reducing human effort and increasing identification performance.”
BackgroundPichler, M., Boreux, V., Klein, A., Schleuning, M., & Hartig, F. (2020) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13329 · Cited by 149 · see citing works
Workflow and convolutional neural network for automated identification of animal sounds
Ruff, Z. J., Lesmeister, D. B., Appel, C. L., & Sullivan, C. M. (2021) · Ecological Indicators · doi:10.1016/j.ecolind.2021.107419 · Cited by 123 · see citing works
Responsible AI for conservation
Wearn, O. R., Freeman, R., & Jacoby, D. M. P. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0022-7 · Cited by 112 · see citing works
Automated Defect Detection and Decision-Support in Gas Turbine Blade Inspection
Aust, J., Shankland, S., Pons, D., Mukundan, R., & Mitrovic, A. (2021) · Aerospace · doi:10.3390/aerospace8020030 · Cited by 66 · see citing works
Palmer, M. S., Huebner, S. E., Willi, M., Fortson, L., & Packer, C. (2021) · Human Computation · doi:10.15346/hc.v8i2.123 · Cited by 20 · see citing works
Liu, S., Xu, A., & Hou, Z. (2026) · Sustainability · doi:10.3390/su18031612
- Authors
- Tabak, M. A., Norouzzadeh, M. S., Wolfson, D. W., Sweeney, S. J., Vercauteren, K. C., Snow, N. P., Halseth, J. M., Di Salvo, P. A., Lewis, J. S., White, M. D., Teton, B., Beasley, J. C., Schlichting, P. E., Boughton, R. K., Wight, B., Newkirk, E. S., Ivan, J. S., Odell, E. A., Brook, R. K., Lukacs, P. M., Moeller, A. K., Mandeville, E. G., Clune, J., & Miller, R. S.
- Affiliations
- University of Wyoming (US)
- United States Department of Agriculture (US)
- Center for Epidemiology and Animal Health United States Department of Agriculture Fort Collins Colorado
- Arizona State University (US)
- Tejon Ranch Conservancy Lebec California
- University of Georgia (US)
- University of Florida (US)
- Colorado Parks and Wildlife (US)
- University of Saskatchewan (CA)
- University of Montana (US)
- Corresponding author
- Tabak, Michael A.
- Miller, Ryan S.
- ORCID
- Tabak, Michael A.
- Norouzzadeh, Mohammad S.
- Wolfson, David W.
- Vercauteren, Kurt C.
- Snow, Nathan P.
- Lewis, Jesse S.
- Beasley, James C.
- Schlichting, Peter E.
- Boughton, Raoul K.
- Wight, Bethany
- Newkirk, Eric S.
- Ivan, Jacob S.
- Odell, Eric A.
- Brook, Ryan K.
- Lukacs, Paul M.
- Moeller, Anna K.
- Mandeville, Elizabeth G.
- Clune, Jeff
- Miller, Ryan S.
- Venue
- Methods in Ecology and Evolution
- Publisher
- Wiley
- Published
- 2018-11-26
- Volume
- 10
- Issue
- 4
- Pages
- 585-590
- ISSN
- 2041-210X (ISSN-L)
- OpenAlex
- W2952113774
- Cited by
- 504
- References
- 34
- Citations by year
- 2026: 64 · 2025: 62 · 2024: 63 · 2023: 75 · 2022: 80 · 2021: 74 · 2020: 56 · 2019: 28 · 2018: 2
- FWCI
- 30.10
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Licence
- Locations
- Methods in Ecology and Evolution (publishedVersion)
- bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion)
- OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
- Lincoln (University of Nebraska) (submittedVersion)
- Funders
- U.S. Department of Energy (DE‐EM0004391)
- U.S. Department of Agriculture
- Colorado Parks and Wildlife
- University of Saskatchewan
- Topics
- Physical Sciences › Environmental Science › Ecology › Wildlife Ecology and Conservation
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Life Sciences › Agricultural and Biological Sciences › Plant Science › Smart Agriculture and AI
- Keywords
- camera trap images, convolutional neural networks, ResNet-18, out-of-distribution generalization, non-invasive sampling, automated image analysis, R package, wildlife monitoring
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
