Computer-Aided Visual Inspection of Glass-Coated Tableware Ceramics for Multi-Class Defect Detection

Quality control procedures in the manufacturing of tableware ceramics require a demanding, monotonous, subjective, and faulty human manual inspection. This paper presents two machine learning strategies and the results of a semi-automated visual inspection of ceramics tableware applied to a private...

Full description

Saved in:
Bibliographic Details
Published in:Applied sciences 2023-11, Vol.13 (21), p.11708
Main Authors: Carvalho, Rafaela, Morgado, Ana C., Gonçalves, João, Kumar, Anil, Rolo, Alberto Gil e Sá, Carreira, Rui, Soares, Filipe
Format: Article
Language:English
Subjects:
Algorithms
Automation
Ceramic industry
Ceramics
Classification
Datasets
defect inspection
Defects
Detectors
Evaluation
International economic relations
Learning strategies
Machine learning
Manufacturing
Medical diagnosis
multi-class
Quality control
Semantics
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Quality control procedures in the manufacturing of tableware ceramics require a demanding, monotonous, subjective, and faulty human manual inspection. This paper presents two machine learning strategies and the results of a semi-automated visual inspection of ceramics tableware applied to a private dataset acquired during the VAICeramics project. In one method, an anomaly detection step was integrated to pre-select possible defective patches before passing through an object detector and defects classifier. In the alternative one, all patches are directly provided to the object detector and then go through the classification phase. Contrary to expectations, the inclusion of the anomaly detector demonstrated a slight reduction in the performance of the pipeline, which may result from error propagation. Regarding the proposed methodology for defect detection, it exhibits average performance in monochromatic images with more than 600 real defects in total, efficiently identifying the most common defect classes in highly reflective surfaces. However, when applied to newly acquired images, the pipeline encounters challenges revealing a lack of generalization ability and experiencing limitations in detecting specific defect classes, due to their appearance and limited available samples used for training. Only two defect types presented high classification performance, namely Dots and Cracked defects.
ISSN:2076-3417
2076-3417
DOI:10.3390/app132111708