基于改进YOLOv3算法的带钢表面缺陷检测
Strip Steel Surface Defect Detection Based on Improved YOLOv3 Algorithm
查看参考文献24篇
|
文摘
|
针对热轧带钢表面缺陷检测中存在的检测速度慢、检测精度低等问题,提出了一种改进的YOLOv3算法模型.使用加权K-means聚类算法来优化确定先验框参数,提高先验框(priors anchor)与特征图层(feature map)的匹配度;同时,调整YOLOv3算法的网络结构,融合浅层特征与深层特征,形成新的大尺度检测图层,提高网络对带钢表面缺陷的检测精度.实验结果表明,改进后的YOLOv3算法在NEU-DET数据集上平均精度均值达到了80%,较原有的YOLOv3算法提高了11%;同时检测速度保持在50fps,优于目前其它深度学习带钢表面缺陷检测算法. |
|
其他语种文摘
|
To solve the problem of slow speed and low accuracy in the surface defect detection of hot rolled strips,an improved YOLOv3 algorithm is proposed.Firstly,the weighting K-means clustering algorithm is put forward to optimize priors anchor's parameters,which can improve the match between priors anchor and feature map.Secondly,the improved network structure of the YOLOv3 algorithm is proposed to improve the detection accuracy,whose shallow features and deep features are combined to form the new large-scale inspection layer.The experiments are carried out on the NEU-DET dataset,the results show that the average accuracy of the improved YOLOv3 algorithm is 80%,which is 11% higher than that of the original algorithm;the detection speed is 50fps,which is faster than other strip surface defect detection algorithms based on deep learning. |
|
来源
|
电子学报
,2020,48(7):1284-1292 【核心库】
|
|
DOI
|
10.3969/j.issn.0372-2112.2020.07.006
|
|
关键词
|
目标检测
;
带钢表面缺陷
;
YOLOv3
;
加权K-means
|
|
地址
|
1.
武汉科技大学, 冶金自动化与检测技术教育部工程研究中心, 湖北, 武汉, 430081
2.
武汉科技大学, 高温材料与炉衬技术国家地方联合工程研究中心, 湖北, 武汉, 430081
3.
武汉科技大学理学院, 湖北, 武汉, 430081
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
0372-2112 |
|
学科
|
自动化技术、计算机技术 |
|
基金
|
国家自然科学基金
|
|
文献收藏号
|
CSCD:6770195
|
参考文献 共
24
共2页
|
|
1.
屈尔庆. 改进的Gabor滤波器带钢表面缺陷显著性检测.
华中科技大学学报(自然科学版),2017,45(10):12-17
|
CSCD被引
8
次
|
|
|
|
|
2.
任海鹏. 基于复杂网络特性的带钢表面缺陷识别.
自动化学报,2011,37(11):1407-1412
|
CSCD被引
5
次
|
|
|
|
|
3.
Zhu S G. A Novel Simple Visual Tracking Algorithm Based on Hashing and Deep Learning.
Chinese Journal of Electronics,2017,26(5):1073-1078
|
CSCD被引
7
次
|
|
|
|
|
4.
Xu K. Application of RNAMlet to surface defect identification of steels.
Optics and Lasers in Engineering,2018,105(6):110-117
|
CSCD被引
7
次
|
|
|
|
|
5.
Song K C. A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects.
Applied Surface Science,2013,285(part_PB):858-864
|
CSCD被引
27
次
|
|
|
|
|
6.
He D. Defect detection of hot rolled steels with a new object detection framework called classification priority network.
Computers & Industrial Engineering,2019,128(3):290-297
|
CSCD被引
12
次
|
|
|
|
|
7.
Fu G Z. A deep-learning-based approach for fast and robust steel surface defects classification.
Optics and Lasers in Engineering,2019,121(10):397-405
|
CSCD被引
11
次
|
|
|
|
|
8.
He Y. Semi-supervised defect classification of steel surface based on multi-training and generative adversarial network.
Optics and Lasers in Engineering,2019,122(11):294-302
|
CSCD被引
13
次
|
|
|
|
|
9.
He D. Surface defect classification of steels with a new semi-supervised learning method.
Optics and Lasers in Engineering,2019,117(6):40-48
|
CSCD被引
17
次
|
|
|
|
|
10.
He Y. An end-to-end steel surface defect detection approach via fusing multiple hierarchical features.
IEEE Transactions on Instrumentation and Measurement,2019,PP(99):1-1
|
CSCD被引
1
次
|
|
|
|
|
11.
姜维. 基于深度学习的场景文字检测综述.
电子学报,2019,47(5):1152-1161
|
CSCD被引
11
次
|
|
|
|
|
12.
张慧. 深度学习在目标视觉检测中的应用进展与展望.
自动化学报,2017,43(8):1289-1305
|
CSCD被引
97
次
|
|
|
|
|
13.
罗会兰. 基于深度学习的视频中人体动作识别进展综述.
电子学报,2019,47(5):1162-1173
|
CSCD被引
23
次
|
|
|
|
|
14.
Ren S Q. Faster R-CNN:towards real-time object detection with region proposal networks.
IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(6):1137-1149
|
CSCD被引
1823
次
|
|
|
|
|
15.
He K M. Mask R-CNN.
Proceedings of the IEEE International Conference on Computer Vision,2017:2961-2969
|
CSCD被引
55
次
|
|
|
|
|
16.
Redmon J. You only look once:Unified,real-time object detection.
Computer Vision and Pattern Recognition,2016:779-788
|
CSCD被引
49
次
|
|
|
|
|
17.
Redmon J. YOLO9000:better,faster,stronger.
Computer Vision and Pattern Recognition,2017:6517-6525
|
CSCD被引
18
次
|
|
|
|
|
18.
Redmon J. YOLOv3:An incremental improvement Computer Vision and Pattern Recognition.
arXiv:1804.02767,2018
|
CSCD被引
479
次
|
|
|
|
|
19.
Liu W. SSD:single shot multibox detector.
Proceedings of the IEEE European Conference on Computer Vision,2016:21-37
|
CSCD被引
6
次
|
|
|
|
|
20.
石杰. 基于改进Mask RCNN和Kinect的服务机器人物品识别系统.
仪器仪表学报,2019,40(4):216-228
|
CSCD被引
14
次
|
|
|
|
|
了解气象卫星更多信息，请访问