SIA OpenIR  > 光电信息技术研究室
基于改进Faster R-CNN的子弹外观缺陷检测
Alternative TitleBullet Appearance Defect Detection Based on Improved Faster Region-Convolutional Neural Network
马晓云1,2,3,4,5; 朱丹1,2,3,4,5; 金晨1,2,4,5; 佟新鑫1,2,4,5
Department光电信息技术研究室
Source Publication激光与光电子学进展
ISSN1006-4125
2019
Volume56Issue:15Pages:1-8
Indexed ByCSCD
CSCD IDCSCD:6573127
Contribution Rank1
Keyword目标检测 Faster R-CNN 子弹外观缺陷 K-means++ 卷积神经网络
Abstract

为了实现子弹外观缺陷自动检测,解决传统机器视觉方法在缺陷检测方面手工设计目标特征耗时和泛化能力差的问题,本文提出基于Faster R-CNN的子弹外观缺陷检测模型。该模型采用卷积神经网络,可以自动提取目标特征,泛化能力强。将检测模型分别与ZFNet、VGG_CNN_M_1024和VGG16结合,实验表明,与VGG16结合的检测模型的检测精度高于其余两种模型方案。针对子弹外观缺陷数据集,本文结合K-means++改进锚框的生成方法,得到基于改进Faster R-CNN的子弹外观缺陷检测模型。结果显示,改进模型在原文算法基础上精度提升至97.75%,速度达到28帧/秒。

Other Abstract

To realize automatic detection of bullet appearance defects and to overcome the limitations associated with traditional machine vision methods, i.e., excessive time required to manually design a target feature and generalization ability is poor in defect detection, we use the K-means++ algorithm to improve the anchor frame generation method and propose a bullet appearance defect detection model based on the improved faster region convolutional neural network (R-CNN).The proposed model uses a CNN that can automatically extract target features and has strong generalization ability. The detection model is combined with ZFNet, VGG_CNN_M_1024, and VGG16,respectively.Results demonstrate that the detection accuracy of the detection model combined with VGG16is higher than the others. The results show that that of the improved model demonstrates 97.75%accuracy and the speed reaches 28frame·s-1.

Language中文
Citation statistics
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/24243
Collection光电信息技术研究室
Corresponding Author马晓云
Affiliation1.中国科学院沈阳自动化研究所
2.中国科学院机器人与智能制造创新研究院
3.中国科学院大学
4.中国科学院光电信息处理研究室
5.辽宁省图像理解与视觉计算重点实验室
Recommended Citation
GB/T 7714
马晓云,朱丹,金晨,等. 基于改进Faster R-CNN的子弹外观缺陷检测[J]. 激光与光电子学进展,2019,56(15):1-8.
APA 马晓云,朱丹,金晨,&佟新鑫.(2019).基于改进Faster R-CNN的子弹外观缺陷检测.激光与光电子学进展,56(15),1-8.
MLA 马晓云,et al."基于改进Faster R-CNN的子弹外观缺陷检测".激光与光电子学进展 56.15(2019):1-8.
Files in This Item: Download All
File Name/Size DocType Version Access License
基于改进Faster_R_CNN的子弹外(2583KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[马晓云]'s Articles
[朱丹]'s Articles
[金晨]'s Articles
Baidu academic
Similar articles in Baidu academic
[马晓云]'s Articles
[朱丹]'s Articles
[金晨]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[马晓云]'s Articles
[朱丹]'s Articles
[金晨]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 基于改进Faster_R_CNN的子弹外观缺陷检测.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.