Abstract:Currently, with the widespread application of embedded technology and the continuous improvement of computational power in mobile terminals, the efficient deployment of algorithms on embedded devices, while maintaining high accuracy and minimizing model size, has become a research hotspot. This paper addresses the challenges of deploying the YOLOv8 algorithm on embedded devices and proposes a novel lightweight object detection algorithm focusing on small object detection. We optimize the model through two key strategies, aiming to achieve lightweight deployment and improve the accuracy of small object detection. Firstly, GhostNet is introduced as the backbone network for YOLOv8 in order to achieve lightweight deployment. By using some cost effective operations to generate redundant feature maps, we not only reduce the number of model parameters while ensuring better detection results, but also improve the speed of the model. Secondly, a new multi-scale attention module is designed to enhance the network's acquisition of crucial information for small targets, which includes a multi-scale fusion attention mechanism and the Soft-NMS algorithm. The multi-scale fusion attention mechanism captures key features of discriminative small targets in the feature map tensor from both spatial and channel dimensions, suppressing non-key information, reducing the impact of complex and unimportant information in the image, enhancing the network model's learning ability for important features of small targets. The Soft-NMS method improves accuracy by significantly reduces false positives in the detection results. To validate the performance of our proposed method, we conducted validation experiments on the PASCAL VOC dataset and evaluated the model's generalization ability on the MS COCO dataset. The experiments results demonstrate that our model achieves a significant improvement in small object detection, with a 5.41% increase in detection accuracy compared to the existing YOLOv8. Meanwhile, FLOPs are reduced by 49.62%, and the number of model parameters is reduced by 48.66%. These results fully confirm the effectiveness of our innovative method in achieving both lightweight deployment and significant efficacy in small object detection tasks.

Lightweight detection method for industrial gas leakage based on improved YOLOv7-tiny

An improved lightweight object detection algorithm for YOLOv5

LW-YOLOv7-Tiny for Remote Sensing Image Object Detection

Light-YOLOv5: A Lightweight Algorithm for Improved YOLOv5 in Complex Fire Scenarios

A Lightweight Underwater Object Detection Model: FL-YOLOV3-TINY

Object Detection Method Based on Lightweight YOLOv4 and Attention Mechanism in Security Scenes.

Object Detection Method Based on Improved YOLO Lightweight Network

LAYN: Lightweight Multi-Scale Attention YOLOv8 Network for Small Object Detection

FL-YOLOv7: A Lightweight Small Object Detection Algorithm in Forest Fire Detection

An improved insulator self-explosion detection method based on group-level pruning for the YOLOv7-tiny algorithm

Lightweight Transmission Line Fault Detection Method Based on Leaner YOLOv7-Tiny

Lightweight Remote Sensing Object Detector Based on YOLOX-Tiny

Lightweight target detection algorithm based on YOLOv4

Fire Detection Method Based on Lightweight YOLOv4

Lightweight and Efficient Tiny-Object Detection Based on Improved YOLOv8n for UAV Aerial Images

Forest Fire Detection Based on Lightweight Yolo

A lightweight fire detection algorithm for small targets based on YOLOv5s

Lightweight Indoor Personnel Detection Algorithm Based on Improved YOLOv4-tiny

F5332SFLFST_YOLOv5s:an improved lightweight aerial photography small target detection algorithm based on YOLOv5s

Lightweight PM-YOLO Network Model for Moving Object Recognition on the Distribution Network Side

Vehicle and Pedestrian Detection Based on Improved YOLOv7-Tiny