Reannotation-Free Automatic Recognition Methods for Empty Camera Trap Images Based on MegaDetector Optimization
Animals · 20.08.2026 · 16 (16) · 2609
DOI: 10.3390/ani16162609 ↗Kurz gefasst
Fachartikel in der Zeitschrift Animals, online erschienen am 20.08.2026. Verfasst von Mei Zhang und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mathematics and Computer Science, Dali University, Dali 671003, China. Die Arbeit nennt 41 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Yunnan Provincial Science and Technology Department University Joint Project, Scientific Research Fund project of the Education Department of Yunnan Province of China.
Abstract (Originalsprache)
Camera traps are widely used in wildlife surveys but often generate numerous empty images due to false triggers, which are time-consuming to filter manually. Microsoft’s MegaDetector (MD) is commonly applied for empty image filtering; however, setting its confidence threshold is challenging because a high threshold increases the false negative rate (FNR), whereas a low threshold raises the false positive rate (FPR), making it difficult to achieve a satisfactory trade-off between the two. To address this dilemma, we proposed three MD-based methods with different performance and computational requirements. For FNR-sensitive users, we proposed MD_CS, which integrated context image similarity to achieve a very low FNR while controlling the FPR. For FPR-sensitive users, we proposed MD_CL, which used MD to generate supervisory signals for training a classification model, thereby reducing the FPR while controlling the FNR. For users requiring both low FNR and low FPR, we proposed MD_CS_CL, an ensemble method that combined the strengths of MD_CS and MD_CL. Experimental results on multiple datasets showed that these three methods achieved a significantly better balance between FNR and FPR compared to the threshold adjustment method of the MD method. Specifically, MD_CS maintained an FNR of 1.8–5.9% while controlling the FPR at 4.9–8.5%; MD_CL kept the FPR at 0.4–2.6% while controlling the FNR at 3.8–8.4%; MD_CS_CL achieved an optimal balance with an FNR of 1.4–4.0% and an FPR of 0.3–0.9%. On a server equipped with an RTX 4090 GPU, the three methods processed 6.7, 5.4, and 4.0 images per second, respectively, and none required manual annotation. Consequently, the proposed methods reduce labor costs, improve work efficiency, and provide reliable technical support for large-scale wildlife monitoring by achieving a superior FNR–FPR trade-off compared to MD’s traditional threshold regulation.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Mathematics and Computer Science, Dali University, Dali 671003, China.
- College of Agricultural and Biological Sciences, Dali University, Dali 671003, China.
- Institute of Eastern-Himalaya Biodiversity Research, Dali University, Dali 671003, China.
- Collaborative Innovation Center for the Biodiversity in the Three Parallel Rivers of China, Dali 671003, China.
- Yunling Black-and-White Snub-Nosed Monkey Observation and Research Station of Yunnan Province, Dali 671003, China.
ORCID-Profile: Wen Xiao, Deng-Qi Yang
Förderung
- National Natural Science Foundation of China
- Yunnan Provincial Science and Technology Department University Joint Project
- Scientific Research Fund project of the Education Department of Yunnan Province of China
- Cultivating Plan Program for the Leader in Science and Technology of Yunnan Province of China
Zitieren
APA
Zhang, M., Yao, J. X., Wu, R. H., Li, X. W., Ren, G. P., Xiao, W., & Yang, D. Q. (2026). Reannotation-Free Automatic Recognition Methods for Empty Camera Trap Images Based on MegaDetector Optimization. Animals, 16(16), 2609. https://doi.org/10.3390/ani16162609
Vancouver
Zhang M, Yao JX, Wu RH, Li XW, Ren GP, Xiao W, et al. Reannotation-Free Automatic Recognition Methods for Empty Camera Trap Images Based on MegaDetector Optimization. Animals. 2026;16(16):2609. doi:10.3390/ani16162609
BibTeX
@article{Zhang2026,
author = {Zhang, Mei and Yao, Jing-Xu and Wu, Rong-Hai and Li, Xiao-Wei and Ren, Guo-Peng and Xiao, Wen and Yang, Deng-Qi},
title = {Reannotation-Free Automatic Recognition Methods for Empty Camera Trap Images Based on MegaDetector Optimization},
journal = {Animals},
year = {2026},
volume = {16},
number = {16},
pages = {2609},
doi = {10.3390/ani16162609}
}
EUVIMED