Visual-Based Defect Detection and Classification Approaches for Industrial Applications-A SURVEY

被引:239
作者
Czimmermann, Tamas [1 ]
Ciuti, Gastone
Milazzo, Mario
Chiurazzi, Marcello
Roccella, Stefano
Oddo, Calogero Maria [1 ]
Dario, Paolo
机构
[1] BioRobot Inst Scuola Super St Anna, I-56025 Pisa, Italy
关键词
defect detection; classification; deep learning; industry; 4.0; survey; AUTOMATED SURFACE INSPECTION; BACKPROPAGATION NEURAL-NETWORK; PATTERN-RECOGNITION; FEATURE-EXTRACTION; TEXTURAL FEATURES; NOVELTY DETECTION; FLAW DETECTION; GABOR FILTERS; COLOR; FOURIER;
D O I
10.3390/s20051459
中图分类号
O65 [分析化学];
学科分类号
070302 ; 081704 ;
摘要
This paper reviews automated visual-based defect detection approaches applicable to various materials, such as metals, ceramics and textiles. In the first part of the paper, we present a general taxonomy of the different defects that fall in two classes: visible (e.g., scratches, shape error, etc.) and palpable (e.g., crack, bump, etc.) defects. Then, we describe artificial visual processing techniques that are aimed at understanding of the captured scenery in a mathematical/logical way. We continue with a survey of textural defect detection based on statistical, structural and other approaches. Finally, we report the state of the art for approaching the detection and classification of defects through supervised and non-supervised classifiers and deep learning.
引用
收藏
页数:25
相关论文
共 221 条
[1]  
ADE F, 1983, P SOC PHOTO-OPT INST, V397, P216, DOI 10.1117/12.935303
[2]  
Ade F., 1984, Seventh International Conference on Pattern Recognition (Cat. No. 84CH2046-1), P428
[3]   Automatic thresholding for defect detection by background histogram mode extents [J].
Aminzadeh, Masoumeh ;
Kurfess, Thomas .
JOURNAL OF MANUFACTURING SYSTEMS, 2015, 37 :83-92
[4]  
[Anonymous], EUFIT 99
[5]  
[Anonymous], FLAW DETECTION WOVEN
[6]  
[Anonymous], P WORKSH TEXT AN MAC
[7]  
[Anonymous], NAT C EM TECHN
[8]  
[Anonymous], BRODATZ TEXTURE DATA
[9]  
[Anonymous], P INT C IM PROC ROCH
[10]  
[Anonymous], LEARNING