Real-Time Recognition Method for 0.8 cm Darning Needles and KR22 Bearings Based on Convolution Neural Networks and Data Increase

被引:24
作者
Yang, Jing [1 ,2 ]
Li, Shaobo [1 ,3 ]
Gao, Zong [3 ]
Wang, Zheng [1 ]
Liu, Wei [2 ,4 ]
机构
[1] Guizhou Univ, Sch Mech Engn, Guiyang 550025, Guizhou, Peoples R China
[2] Oklahoma State Univ, Sch Elect & Comp Engn, Stillwater, OK 74078 USA
[3] Guizhou Univ, Guizhou Prov Key Lab Publ Big Data, Guiyang 550025, Guizhou, Peoples R China
[4] Xian Polytech Univ, Sch Elect & Informat, Xian 710048, Shaanxi, Peoples R China
来源
APPLIED SCIENCES-BASEL | 2018年 / 8卷 / 10期
基金
中国国家自然科学基金;
关键词
data increase; object recognition; precision parts; 0.8cm darning needle; KR22; bearing; CLASSIFICATION; MODEL;
D O I
10.3390/app8101857
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
The complexity of the background and the similarities between different types of precision parts, especially in the high-speed movement of conveyor belts in complex industrial scenes, pose immense challenges to the object recognition of precision parts due to diversity in illumination. This study presents a real-time object recognition method for 0.8 cm darning needles and KR22 bearing machine parts under a complex industrial background. First, we propose an image data increase algorithm based on directional flip, and we establish two types of dataset, namely, real data and increased data. We focus on increasing recognition accuracy and reducing computation time, and we design a multilayer feature fusion network to obtain feature information. Subsequently, we propose an accurate method for classifying precision parts on the basis of non-maximal suppression, and then form an improved You Only Look Once (YOLO) V3 network. We implement this method and compare it with models in our real-time industrial object detection experimental platform. Finally, experiments on real and increased datasets show that the proposed method outperforms the YOLO V3 algorithm in terms of recognition accuracy and robustness.
引用
收藏
页数:18
相关论文
共 30 条
[1]  
[Anonymous], 2018, IEEE T INF FOREN SEC, DOI DOI 10.1109/TIFS.2018.2812196
[2]   ENHANCING CONSTRUCTIVE NEURAL NETWORK PERFORMANCE USING FUNCTIONALLY EXPANDED INPUT DATA [J].
Bertini Junior, Joao Roberto ;
Nicoletti, Maria do Carmo .
JOURNAL OF ARTIFICIAL INTELLIGENCE AND SOFT COMPUTING RESEARCH, 2016, 6 (02) :119-131
[3]   Densely Connected Convolutional Neural Network for Multi-purpose Image Forensics under Anti-forensic Attacks [J].
Chen, Yifang ;
Kang, Xiangui ;
Wang, Z. Jane ;
Zhan, Qiong .
PROCEEDINGS OF THE 6TH ACM WORKSHOP ON INFORMATION HIDING AND MULTIMEDIA SECURITY (IH&MMSEC'18), 2018, :91-96
[4]   Automatic moth detection from trap images for pest management [J].
Ding, Weiguang ;
Taylor, Graham .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2016, 123 :17-28
[5]   Changes in the Complexity of Heart Rate Variability with Exercise Training Measured by Multiscale Entropy-Based Measurements [J].
Fazan, Frederico Sassoli ;
Brognara, Fernanda ;
Fazan Junior, Rubens ;
Murta Junior, Luiz Otavio ;
Virgilio Silva, Luiz Eduardo .
ENTROPY, 2018, 20 (01)
[6]  
Ganovska B, 2016, INDIAN J ENG MATER S, V23, P279
[7]   A recurrent neural network based health indicator for remaining useful life prediction of bearings [J].
Guo, Liang ;
Li, Naipeng ;
Jia, Feng ;
Lei, Yaguo ;
Lin, Jing .
NEUROCOMPUTING, 2017, 240 :98-109
[8]  
He T, 2017, 2017 INTERNATIONAL CONFERENCE ON MECHANICAL, SYSTEM AND CONTROL ENGINEERING (ICMSC), P63, DOI 10.1109/ICMSC.2017.7959444
[9]   Fast Detection of Head Colliding Shapes on Automobile Parts [J].
Inui, Masatomo ;
Umezu, Nobuyuki .
JOURNAL OF ADVANCED MECHANICAL DESIGN SYSTEMS AND MANUFACTURING, 2013, 7 (05) :818-826
[10]   Deep normalized convolutional neural network for imbalanced fault classification of machinery and its understanding via visualization [J].
Jia, Feng ;
Lei, Yaguo ;
Lu, Na ;
Xing, Saibo .
MECHANICAL SYSTEMS AND SIGNAL PROCESSING, 2018, 110 :349-367