•  
  •  
 

Abstract

The proliferation of unmanned aerial vehicles (UAVs) has necessitated the development of sophisticated object detection algorithms capable of handling the unique challenges posed by aerial imagery. Traditional detection methods often struggle with small object sizes, dense distributions, and complex backgrounds characteristic of UAV-captured scenes. This research presents EYOLOv8-MSAFF (Enhanced YOLOv8 with Multi-Scale Attention and Feature Fusion), a novel deep learning architecture specifically engineered for superior performance in UAV-based object detection tasks. The proposed methodology integrates four innovative components: a Hybrid Spatial-Channel Attention Mechanism (HSCAM) that processes attention information in parallel rather than sequentially, an Adaptive Multi-Scale Feature Fusion Module (AMSFFM) that dynamically weights multi-scale features based on contextual relevance, a Dilated Convolutional Block Attention Module (DCBAM) that captures multi-scale contextual information through varying dilation rates, and a Progressive Feature Refinement Network (PFRN) that iteratively enhances feature representations through multiple refinement stages. Extensive experimental validation conducted on two challenging datasets demonstrates the exceptional performance of the proposed approach. On the VisDrone2019 dataset, EYOLOv8-MSAFF achieves 87.3% mAP@0.5 and 64.2% mAP@0.5:0.95, representing substantial improvements of 6.1% and 7.1% respectively over the baseline YOLOv8l architecture. The method exhibits particularly remarkable performance in small object detection, achieving 52.7% AP_S, which constitutes an 18.2% improvement over the baseline. Similarly, on the DOTA v2.0 dataset, the proposed method attains 85.7% mAP@0.5 and 58.9% mAP@0.5:0.95, surpassing existing state-of-the-art approaches by significant margins. Comprehensive ablation studies validate the effectiveness of each proposed component, while computational analysis demonstrates that the method maintains practical inference speeds of 33.4 FPS despite the incorporation of sophisticated attention mechanisms.

Download

Share

COinS