To address challenges in UAV aerial images such as a high proportion of small objects, complex backgrounds, significant scale variations, and limited compu-ting resources, this study proposes an improved object detection algorithm based on YOLOv11n—namely, IRE-YOLO. In the backbone network, the algo-rithm introduces the inverted residual mobile block (iRMB) attention mecha-nism to enhance the model's ability to extract features of small objects in com-plex backgrounds. It replaces some standard convolutions with depthwise sepa-rable convolutions (DSConv) to reduce the number of parameters while main-taining feature expression capability. In the neck network, an efficient upsam-pling block (EUCB) is designed to recover detailed information through hierar-chical feature reconstruction, thereby improving the representation quality of small objects' edges and textures. Experimental results on the DOTA dataset show that IRE-YOLO outperforms mainstream lightweight models in key metrics including precision (P), recall (R), mAP@50, and mAP@50:95, achieving 69.4%, 59.0%, 61.0%, and 34.2% respectively. Meanwhile, the model's parameter count is reduced to 8.7MB, realizing an effective balance between accuracy and light-weight design. The research demonstrates that IRE-YOLO has high practical val-ue in small object detection tasks for UAV aerial images and provides a feasible solution for real-time deployment on embedded platforms.
Building similarity graph...
Analyzing shared references across papers
Loading...
Yang Liu
Shenyang Aerospace University
Tianjin University of Technology and Education
Building similarity graph...
Analyzing shared references across papers
Loading...
Yang Liu (Wed,) studied this question.
synapsesocial.com/papers/68de68e583cbc991d0a20f18 — DOI: https://doi.org/10.63313/ajet.9007