Surface Defect Detection of Steel Strip at Low Resolution Based on SAC-YOLOv5

计算机科学 卷积神经网络 人工智能 特征(语言学) 块(置换群论) 模式识别(心理学) 推论 人工神经网络 曲面(拓扑) 算法 数学 几何学 语言学 哲学
作者
changxin rui,Zhantao Wu,C. Liu,Baoqing Li,Junsheng Cheng
出处
期刊:Measurement Science and Technology [IOP Publishing]
标识
DOI:10.1088/1361-6501/ad9e25
摘要

Abstract Surface defects are common occurrences in the production process of strip steel. The development of automatic and efficient intelligent detection algorithms for strip steel is crucial for enhancing product quality and operational safety. While deep learning-based defect detection techniques have achieved satisfactory accuracy when applied to high-resolution images, obtaining a sufficient number of high-resolution images in practical engineering scenarios is challenging. The degradation of image quality often results in a significant decline in the performance of existing detection techniques. To address these challenges, this paper proposes SAC-YOLOv5-based surface defect detection model specifically designed for low-resolution strip steel images. SAC, SIoU based K-Means++, Asymmetric Convolutional Neural Networks and Composite Down-sampling Feature Fusion Block. Firstly, we introduce an anchor boxes clustering algorithm called Shape-IoU based K-Means++ (SIoU based K-Means++) to enhance the efficiency of anchor boxes regression. Secondly, we construct an Asymmetric Convolutional Neural Networks (ACNN) for multi-level feature extraction. Utilizing reparameterization techniques, we reduce the inference resources required by the model. Additionally, we propose a Composite Down-sampling Feature Fusion Block (CDFFB) which enhances key texture information and improves the model's nonlinear fitting ability. Experimental analysis using NEU-DET surface defect data demonstrates the superiority of the proposed model in processing low-resolution strip steel surface defect images. In comparsion to YOLOv8, YOLOv7, YOLOX, YOLOv3SPP, CenterNet and Faster RCNN, SAC-YOLOv5 outperforms all other models in terms of both speed and accuracy.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

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
刚刚
chen完成签到,获得积分10
刚刚
Hello应助ylw采纳,获得10
刚刚
1秒前
ChemistryZyh发布了新的文献求助10
1秒前
wensir发布了新的文献求助10
1秒前
端庄千琴完成签到,获得积分10
1秒前
heavennew完成签到,获得积分10
2秒前
3秒前
眼睛大樱桃完成签到,获得积分10
3秒前
Yuantian发布了新的文献求助10
4秒前
学吗你完成签到 ,获得积分10
4秒前
御青白少发布了新的文献求助10
5秒前
无尽夏完成签到,获得积分10
5秒前
Rylee发布了新的文献求助10
7秒前
7秒前
无私的念文完成签到 ,获得积分10
8秒前
充电宝应助Yuantian采纳,获得10
9秒前
水水完成签到,获得积分10
10秒前
sskr发布了新的文献求助10
10秒前
15327432191完成签到 ,获得积分10
11秒前
酷波er应助果汁采纳,获得10
11秒前
善学以致用应助程公子采纳,获得10
11秒前
海阔天空发布了新的文献求助10
11秒前
ChemistryZyh完成签到,获得积分10
12秒前
wensir完成签到,获得积分10
14秒前
斯文败类应助Rylee采纳,获得10
15秒前
养不熟的野猫完成签到,获得积分10
15秒前
sskr完成签到,获得积分10
15秒前
高文强完成签到,获得积分10
16秒前
17秒前
我是老大应助liu采纳,获得10
17秒前
领导范儿应助小熊软糖采纳,获得10
17秒前
华仔应助kevin采纳,获得10
18秒前
19秒前
20秒前
晴朗葡萄发布了新的文献求助30
20秒前
bkagyin应助冷艳的冬萱采纳,获得10
20秒前
21秒前
文献啊文献完成签到,获得积分10
22秒前
高分求助中
【提示信息,请勿应助】关于scihub 10000
Les Mantodea de Guyane: Insecta, Polyneoptera [The Mantids of French Guiana] 3000
徐淮辽南地区新元古代叠层石及生物地层 3000
The Mother of All Tableaux: Order, Equivalence, and Geometry in the Large-scale Structure of Optimality Theory 3000
Global Eyelash Assessment scale (GEA) 1000
Picture Books with Same-sex Parented Families: Unintentional Censorship 550
Research on Disturbance Rejection Control Algorithm for Aerial Operation Robots 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 4038569
求助须知:如何正确求助?哪些是违规求助? 3576279
关于积分的说明 11374944
捐赠科研通 3305979
什么是DOI,文献DOI怎么找? 1819354
邀请新用户注册赠送积分活动 892698
科研通“疑难数据库(出版商)”最低求助积分说明 815048