Drones are widely used for wildlife monitoring. Deep learning algorithms are key to the success of monitoring wildlife with drones, although they face the problem of detecting small targets. To solve this problem, we have introduced the SE-YOLO model, which incorporates a channel self-attention mechanism into the advanced real-time object detection algorithm YOLOv7, enabling the model to perform effectively on small targets. However, there is another barrier; the lack of publicly available UAV wildlife aerial datasets hampers research on UAV wildlife monitoring algorithms. To fill this gap, we present a large-scale, multi-class, high-quality dataset called WAID (Wildlife Aerial Images from Drone), which contains 14,375 UAV aerial images from different environmental conditions, covering six wildlife species and multiple habitat types. We conducted a statistical analysis experiment, an algorithm detection comparison experiment, and a dataset generalization experiment. The statistical analysis experiment demonstrated the dataset characteristics both quantitatively and intuitively. The comparison and generalization experiments compared different types of advanced algorithms as well as the SE-YOLO method from the perspective of the practical application of UAVs for wildlife monitoring. The experimental results show that WAID is suitable for the study of wildlife monitoring algorithms for UAVs, and SE-YOLO is the most effective in this scenario, with a mAP of up to 0.983. This study brings new methods, data, and inspiration to the field of wildlife monitoring by UAVs.

Grassland ecosystems are vital for global biodiversity, yet traditional wildlife monitoring methods are often labor-intensive and costly. Although drone-based aerial surveys provide a scalable alternative, they face significant challenges such as detecting extremely small targets, handling complex backgrounds, and operating under strict computational constraints. To address these issues, this study proposes IECA-YOLOv7, a lightweight detection model that incorporates three key innovations: an Improved Efficient Channel Attention (IECA) module for enhanced feature representation, a content-aware CARAFE upsampling operator for improved detail recovery, and a Normalized Wasserstein Distance (NWD) loss function for robust small-target localization. Evaluated on a dedicated grassland wildlife dataset (GWAID), the model achieves a mAP@0.5 of 86.6% and a mAP@0.5:0.95 of 47.2%, outperforming the baseline YOLOv7-tiny by 2.9% in Precision and 1.8% in Recall. Furthermore, it surpasses non-YOLO architectures such as RetinaNet, EfficientDet-D0, and DETR by significant margins, demonstrating superior performance in small-object detection under complex conditions. Cross-dataset validation on VisDrone, CARPK, and DOTA demonstrates a strong generalization capability. With a model size under 5 MB, IECA-YOLOv7 effectively balances accuracy and efficiency, offering a practical solution for real-time wildlife monitoring via drones under challenging environmental constraints such as variable lighting, occlusion, and limited computational resources, thereby supporting broader conservation efforts.