A Method of Defect Detection for Focal Hard Samples PCB Based on Extended FPN Model

特征(语言学) 棱锥(几何) 计算机科学 人工智能 目标检测 图层(电子) 深度学习 算法 模式识别(心理学) 数学 几何学 语言学 哲学 有机化学 化学
作者
Cui‐Jin Li,Zhong Qu,Shiyan Wang,Kang-Hua Bao,Shengye Wang
出处
期刊:IEEE Transactions on Components, Packaging and Manufacturing Technology [Institute of Electrical and Electronics Engineers]
卷期号:12 (2): 217-227 被引量:34
标识
DOI:10.1109/tcpmt.2021.3136823
摘要

Suffering from the diversity, complexity, and miniaturization of printed circuit board (PCB) defects, traditional detection methods are difficult to detect. Despite object detection has made significant advances based on deep neural networks, it remains a challenge to focus on small objects. We address this challenge by allowing multiscale fusion. We introduce a PCB defect detection algorithm based on extended feature pyramid network model in this article. The backbone is constructed by part of ResNet-101, in order to accurately locate and identify small objects, this article constructs a feature layer, which integrates high-level semantic information and low-level geometric information. Based on feature pyramid networks (FPN) network structure, using $1\times1$ convolution lateral fusion of the previous semantic information, the fused features use $3\times3$ convolution to obtain the final feature layer. The problem that PCB defects are difficult to classify is considered, the focal loss function is introduced. To reduce over-fitting in the training process, the original data are enhanced using image clipping and rotation. Through the quantitative analysis on PCB defect dataset, these results are the best to be used in fused low-level feature layer for detection of the mean average precision (mAP). This is 96.2% on the public PCB dataset, which is surpassing the state-of-the-art methods.
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
刚刚
刚刚
刘二宝发布了新的文献求助20
刚刚
甜甜发带完成签到,获得积分20
刚刚
方大完成签到,获得积分10
1秒前
2秒前
louxiaohan完成签到,获得积分10
2秒前
2秒前
Liu发布了新的文献求助10
3秒前
sunny发布了新的文献求助10
3秒前
3秒前
Yixiaofei发布了新的文献求助10
3秒前
4秒前
4秒前
Wu发布了新的文献求助10
5秒前
rose发布了新的文献求助10
5秒前
静静完成签到,获得积分10
5秒前
12w发布了新的文献求助10
5秒前
李爱国应助甜甜发带采纳,获得10
6秒前
张宇姝完成签到,获得积分10
6秒前
Bran发布了新的文献求助10
6秒前
温热如河完成签到,获得积分20
6秒前
嘚嘚完成签到,获得积分10
7秒前
shilong.yang发布了新的文献求助10
7秒前
7秒前
斯文败类应助yyyyyge采纳,获得10
7秒前
qian完成签到,获得积分10
8秒前
8秒前
8秒前
8秒前
打打应助cslghe采纳,获得10
9秒前
hhuajw完成签到,获得积分10
9秒前
9秒前
9秒前
甜美板栗完成签到,获得积分10
9秒前
hmf1995完成签到 ,获得积分10
10秒前
失眠朋友发布了新的文献求助10
10秒前
ptiying发布了新的文献求助10
10秒前
轻松闭月完成签到 ,获得积分20
11秒前
高分求助中
Picture Books with Same-sex Parented Families: Unintentional Censorship 700
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
不知道标题是什么 500
Indomethacinのヒトにおける経皮吸収 400
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 370
Effective Learning and Mental Wellbeing 300
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 3974602
求助须知:如何正确求助?哪些是违规求助? 3519026
关于积分的说明 11196787
捐赠科研通 3255127
什么是DOI,文献DOI怎么找? 1797693
邀请新用户注册赠送积分活动 877100
科研通“疑难数据库(出版商)”最低求助积分说明 806130