Abstract:Object detection in unmanned aerial vehicle (UAV) imagery is a meaningful foundation in various research domains. However, UAV imagery poses unique challenges, including large image sizes, small sizes detection objects, dense distribution, overlapping instances, and insufficient lighting impacting the effectiveness of object detection. In this article, we propose Drone-YOLO, a series of multi-scale UAV image object detection algorithms based on the YOLOv8 model, designed to overcome the specific challenges associated with UAV image object detection. To address the issues of large scene sizes and small detection objects, we introduce improvements to the neck component of the YOLOv8 model. Specifically, we employ a three-layer PAFPN structure and incorporate a detection head tailored for small-sized objects using large-scale feature maps, significantly enhancing the algorithm’s capability to detect small-sized targets. Furthermore, we integrate the sandwich-fusion module into each layer of the neck’s up–down branch. This fusion mechanism combines network features with low-level features, providing rich spatial information about the objects at different layer detection heads. We achieve this fusion using depthwise separable evolution, which balances parameter costs and a large receptive field. In the network backbone, we employ RepVGG modules as downsampling layers, enhancing the network’s ability to learn multi-scale features and outperforming traditional convolutional layers. The proposed Drone-YOLO methods have been evaluated in ablation experiments and compared with other state-of-the-art approaches on the VisDrone2019 dataset. The results demonstrate that our Drone-YOLO (large) outperforms other baseline methods in the accuracy of object detection. Compared to YOLOv8, our method achieves a significant improvement in mAP0.5 metrics, with a 13.4% increase on the VisDrone2019-test and a 17.40% increase on the VisDrone2019-val. Additionally, the parameter-efficient Drone-YOLO (tiny) with only 5.25 M parameters performs equivalently or better than the baseline method with 9.66M parameters on the dataset. These experiments validate the effectiveness of the Drone-YOLO methods in the task of object detection in drone imagery.

ViT-YOLO:Transformer-Based YOLO for Object Detection

TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarios

TPH-YOLOv5++: Boosting Object Detection on Drone-Captured Scenarios with Cross-Layer Asymmetric Transformer

Drone-YOLO: An Efficient Neural Network Method for Target Detection in Drone Images

An Object Detection Method Based on Improved YOLOX

YOLO-UAV: Object Detection Method of Unmanned Aerial Vehicle Imagery Based on Efficient Multi-Scale Feature Fusion

FE-YOLOv5: Improved YOLOv5 Network for Multi-scale Drone-Captured Scene Detection

LV-YOLOv5: A Light-Weight Object Detector of Vit on Drone-captured Scenarios

DB-YOLOv5: A UAV Object Detection Model Based on Dual Backbone Network for Security Surveillance

YOLO-Drone: A Scale-Aware Detector for Drone Vision

MS-YOLO: integration-based multi-subnets neural network for object detection in aerial images

DMA-YOLO: Multi-Scale Object Detection Method with Attention Mechanism for Aerial Images

Improved YOLOv7 Based on Transformer for Object Detection in UAV-Captured Images

DM-YOLOX Aerial Object Detection Method with Intensive Attention Mechanism

LAM-YOLO: Drones-based Small Object Detection on Lighting-Occlusion Attention Mechanism YOLO

Towards a High-Performance Object Detector: Insights from Drone Detection Using ViT and CNN-based Deep Learning Models

HIC-YOLOv5: Improved YOLOv5 For Small Object Detection

IV-YOLO: A Lightweight Dual-Branch Object Detection Network

BGF-YOLOv10: Small Object Detection Algorithm from Unmanned Aerial Vehicle Perspective Based on Improved YOLOv10

MRT-YOLO: A Fine-Grained Feature-Based Method for Object Detection