SOD-YOLOv8 -- Enhancing YOLOv8 for Small Object Detection in Traffic Scenes

计算机视觉 对象(语法) 计算机科学 目标检测 人工智能 模式识别(心理学)
作者
Boshra Khalili,Andrew W. Smyth
出处
期刊:Cornell University - arXiv [Cornell University]
被引量:4
标识
DOI:10.48550/arxiv.2408.04786
摘要

Object detection as part of computer vision can be crucial for traffic management, emergency response, autonomous vehicles, and smart cities. Despite significant advances in object detection, detecting small objects in images captured by distant cameras remains challenging due to their size, distance from the camera, varied shapes, and cluttered backgrounds. To address these challenges, we propose Small Object Detection YOLOv8 (SOD-YOLOv8), a novel model specifically designed for scenarios involving numerous small objects. Inspired by Efficient Generalized Feature Pyramid Networks (GFPN), we enhance multi-path fusion within YOLOv8 to integrate features across different levels, preserving details from shallower layers and improving small object detection accuracy. Also, A fourth detection layer is added to leverage high-resolution spatial information effectively. The Efficient Multi-Scale Attention Module (EMA) in the C2f-EMA module enhances feature extraction by redistributing weights and prioritizing relevant features. We introduce Powerful-IoU (PIoU) as a replacement for CIoU, focusing on moderate-quality anchor boxes and adding a penalty based on differences between predicted and ground truth bounding box corners. This approach simplifies calculations, speeds up convergence, and enhances detection accuracy. SOD-YOLOv8 significantly improves small object detection, surpassing widely used models in various metrics, without substantially increasing computational cost or latency compared to YOLOv8s. Specifically, it increases recall from 40.1\% to 43.9\%, precision from 51.2\% to 53.9\%, $\text{mAP}_{0.5}$ from 40.6\% to 45.1\%, and $\text{mAP}_{0.5:0.95}$ from 24\% to 26.6\%. In dynamic real-world traffic scenes, SOD-YOLOv8 demonstrated notable improvements in diverse conditions, proving its reliability and effectiveness in detecting small objects even in challenging environments.

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
123完成签到,获得积分10
1秒前
芽芽配茄子完成签到,获得积分10
1秒前
123完成签到,获得积分10
1秒前
YPHCC发布了新的文献求助10
1秒前
1秒前
111完成签到,获得积分10
1秒前
山止川行完成签到 ,获得积分10
2秒前
隐形曼青的应助被锋zai采纳,获得10
2秒前
2秒前
记得吃早饭完成签到,获得积分10
2秒前
tfsn20完成签到,获得积分0
2秒前
Lucky晨完成签到,获得积分10
2秒前
绵马紫萁完成签到,获得积分10
2秒前
lisier完成签到,获得积分10
2秒前
3秒前
马嘉祺完成签到,获得积分10
3秒前
科研小白完成签到,获得积分10
4秒前
热心的山水完成签到,获得积分10
4秒前
灵巧的TL完成签到 ,获得积分10
4秒前
4秒前
4秒前
健忘飞珍完成签到,获得积分10
4秒前
水土洼完成签到,获得积分10
4秒前
芭乐发布了新的文献求助10
4秒前
Jelly完成签到,获得积分10
4秒前
nixuan完成签到,获得积分10
5秒前
大花花完成签到,获得积分10
5秒前
molihuakai的应助被工具人杨肥肥采纳,获得10
5秒前
Cjl10610的应助被YFFDSG采纳,获得10
5秒前
鲸落小小完成签到,获得积分10
5秒前
昭昭昭昭完成签到,获得积分10
5秒前
汉堡包的应助被优雅莞采纳,获得10
6秒前
小猫炸毛完成签到,获得积分10
6秒前
薯条完成签到,获得积分10
6秒前
6秒前
小二郎的应助被迷人葶采纳,获得10
7秒前
嗯我就不说完成签到,获得积分10
7秒前
tong发布了新的文献求助10
7秒前
8秒前
8秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
Social Psychology in the Real World 600
八旗制度の研究 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8180275
求助须知:如何正确求助?哪些是违规求助? 9570366
关于积分的说明 21810725
捐赠科研通 7776380
什么是DOI,文献DOI怎么找? 3433277
关于科研通互助平台的介绍 2557483
邀请新用户注册赠送积分活动 2479042