This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
DMN-YOLO: A Lightweight Small-Object Detector for Multi-Species Animal Detection in UAV Grassland Imagery
1
School of Artificial Intelligence, Anhui Agricultural University, Hefei 230036, China
2
Key Laboratory of Agricultural Sensors, Ministry of Agriculture and Rural Affairs, Anhui Agricultural University, Hefei 230036, China
*
Author to whom correspondence should be addressed.
Animals 2026, 16(11), 1643; https://doi.org/10.3390/ani16111643
Submission received: 30 April 2026
/
Revised: 21 May 2026
/
Accepted: 23 May 2026
/
Published: 27 May 2026
(This article belongs to the Section Animal System and Management)
Simple Summary
Monitoring animals in open grasslands is important for grazing management, wildlife protection, and ecological conservation, but traditional field surveys are often slow, labor-intensive, and limited by terrain. Drone images provide a more efficient way to observe animals over large areas, but animals often appear very small in aerial images and can be difficult to distinguish from grass, shadows, and other background objects. This study developed a lightweight computer-based method to detect multiple animal species, including cattle, sheep, and wild animals, in drone images of grassland scenes. The method was designed to improve the detection of small animals while keeping the model compact enough for practical use. The results showed that the proposed method correctly identified most animal targets and performed better than the original model while using a smaller model size. This study provides a practical tool for animal monitoring, grazing management, ecological patrols, and intelligent grassland conservation.
Abstract
To meet the requirements of accurate multi-class animal detection and model lightweighting in UAV-based grazing monitoring, this study presents DMN-YOLO, an efficient detector built upon YOLO11n. In particular, a lightweight downsampling module, DSDown, is introduced to alleviate the loss of detailed features of tiny targets during downsampling under complex grassland backgrounds, thereby improving the preservation of edge, texture, and local structural information. Meanwhile, a MACFPN multi-scale feature fusion structure is designed to handle large scale variations and feature confusion among multiple animal targets, enhancing cross-scale feature interaction and background suppression for better small-target representation. In addition, NWDR Loss combines CIoU geometric constraints, normalized Wasserstein distance, and an adaptive weighting strategy to improve overall stability and localization accuracy of small-target bounding box regression. Results indicate that DMN-YOLO attains 93.6% precision, 89.9% recall, and 95.8% mAP@0.5 on the UAV animal detection dataset. Compared with YOLO11n, it reduces the parameter count by 35.7% while lowering the model size by 29.3%. These results show that DMN-YOLO effectively reduces model complexity while maintaining strong detection performance, demonstrating good potential for practical field deployment.
