Comparison of a Computer Vision Model to a Human Observer in Detecting African Mammals in Camera Trap Images within a Safari Park
Bibliographic record
Abstract
Remote monitoring technologies are increasingly utilized in animal research for their capacity to enhance data collection efficiency. However, they present challenges, and as such researchers have resorted to utilizing deep learning to automatically classify acquired data therefore expediting the review process. While this practice is common in field studies it has been less adopted in zoo monitoring. In this paper we deploy the YOLOv10x model to monitor four species at Knowsley Safari in the UK: African lions ( Panthera leo ), Southern white rhino ( Ceratotherium simum simum ), Grevy’s zebra ( Equus grevyi ) and Olive baboons ( Papio anubis ). Camera trap images were processed and classified using the Conservation AI desktop application. The raw images were saved to facilitate the comparative analysis of the models’ predictions against the findings of human observed images. Processing time for both methods was compared using a subset of 3015 images with Conservation AI, reducing the time required to classify the images by 82% compared to a human analyst. Confusion matrix results showed high accuracy rates for all four species (>0.90). Analysis of count data showed significant differences in three species, where the human observer recorded more observations of each than Conservation AI (lion, rhino, baboon p<0.005). However, no significant difference was seen in zebra (p>0.05). A strong positive correlation in count data between both methodologies was seen in all species; baboon (rho=0.955, p<0.005), lion (rho = 0.969, p<0.005), rhino (rho=0.887, p<0.005) and zebra (rho=0.843, p<0.005). This study highlights the potential for these technologies as a monitoring system in zoos.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".