A Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry

Manufacturers are eager to replace the human inspector with automatic inspection systems to improve the competitive advantage by means of quality. However, some manufacturers have failed to apply the traditional vision system because of constraints in data acquisition and feature extraction. In this...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Donggyun Im, Sangkyu Lee, Homin Lee, Byungguan Yoon, Fayoung So, Jongpil Jeong
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	surface inspection deep learning industrial control process automation CLAHE YOLO Chemical technology TP1-1185 Chemistry QD1-999
Acceso en línea:	https://doaj.org/article/9cf7e32a84c64e71944e3f7211716fb1
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:9cf7e32a84c64e71944e3f7211716fb1
record_format	dspace
spelling	oai:doaj.org-article:9cf7e32a84c64e71944e3f7211716fb12021-11-25T18:50:11ZA Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry10.3390/pr91118952227-9717https://doaj.org/article/9cf7e32a84c64e71944e3f7211716fb12021-10-01T00:00:00Zhttps://www.mdpi.com/2227-9717/9/11/1895https://doaj.org/toc/2227-9717Manufacturers are eager to replace the human inspector with automatic inspection systems to improve the competitive advantage by means of quality. However, some manufacturers have failed to apply the traditional vision system because of constraints in data acquisition and feature extraction. In this paper, we propose an inspection system based on deep learning for a tampon applicator producer that uses the applicator’s structural characteristics for data acquisition and uses state-of-the-art models for object detection and instance segmentation, YOLOv4 and YOLACT for feature extraction, respectively. During the on-site trial test, we experienced some False-Positive (FP) cases and found a possible Type I error. We used a data-centric approach to solve the problem by using two different data pre-processing methods, the Background Removal (BR) and Contrast Limited Adaptive Histogram Equalization (CLAHE). We have experimented with analyzing the effect of the methods on the inspection with the self-created dataset. We found that CLAHE increased Recall by 0.1 at the image level, and both CLAHE and BR improved Precision by 0.04–0.06 at the bounding box level. These results support that the data-centric approach might improve the detection rate. However, the data pre-processing techniques deteriorated the metrics used to measure the overall performance, such as F1-score and Average Precision (AP), even though we empirically confirmed that the malfunctions improved. With the detailed analysis of the result, we have found some cases that revealed the ambiguity of the decisions caused by the inconsistency in data annotation. Our research alerts AI practitioners that validating the model based only on the metrics may lead to a wrong conclusion.Donggyun ImSangkyu LeeHomin LeeByungguan YoonFayoung SoJongpil JeongMDPI AGarticlesurface inspectiondeep learningindustrial controlprocess automationCLAHEYOLOChemical technologyTP1-1185ChemistryQD1-999ENProcesses, Vol 9, Iss 1895, p 1895 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	surface inspection deep learning industrial control process automation CLAHE YOLO Chemical technology TP1-1185 Chemistry QD1-999
spellingShingle	surface inspection deep learning industrial control process automation CLAHE YOLO Chemical technology TP1-1185 Chemistry QD1-999 Donggyun Im Sangkyu Lee Homin Lee Byungguan Yoon Fayoung So Jongpil Jeong A Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry
description	Manufacturers are eager to replace the human inspector with automatic inspection systems to improve the competitive advantage by means of quality. However, some manufacturers have failed to apply the traditional vision system because of constraints in data acquisition and feature extraction. In this paper, we propose an inspection system based on deep learning for a tampon applicator producer that uses the applicator’s structural characteristics for data acquisition and uses state-of-the-art models for object detection and instance segmentation, YOLOv4 and YOLACT for feature extraction, respectively. During the on-site trial test, we experienced some False-Positive (FP) cases and found a possible Type I error. We used a data-centric approach to solve the problem by using two different data pre-processing methods, the Background Removal (BR) and Contrast Limited Adaptive Histogram Equalization (CLAHE). We have experimented with analyzing the effect of the methods on the inspection with the self-created dataset. We found that CLAHE increased Recall by 0.1 at the image level, and both CLAHE and BR improved Precision by 0.04–0.06 at the bounding box level. These results support that the data-centric approach might improve the detection rate. However, the data pre-processing techniques deteriorated the metrics used to measure the overall performance, such as F1-score and Average Precision (AP), even though we empirically confirmed that the malfunctions improved. With the detailed analysis of the result, we have found some cases that revealed the ambiguity of the decisions caused by the inconsistency in data annotation. Our research alerts AI practitioners that validating the model based only on the metrics may lead to a wrong conclusion.
format	article
author	Donggyun Im Sangkyu Lee Homin Lee Byungguan Yoon Fayoung So Jongpil Jeong
author_facet	Donggyun Im Sangkyu Lee Homin Lee Byungguan Yoon Fayoung So Jongpil Jeong
author_sort	Donggyun Im
title	A Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry
title_short	A Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry
title_full	A Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry
title_fullStr	A Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry
title_full_unstemmed	A Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry
title_sort	data-centric approach to design and analysis of a surface-inspection system based on deep learning in the plastic injection molding industry
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/9cf7e32a84c64e71944e3f7211716fb1
work_keys_str_mv	AT donggyunim adatacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT sangkyulee adatacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT hominlee adatacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT byungguanyoon adatacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT fayoungso adatacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT jongpiljeong adatacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT donggyunim datacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT sangkyulee datacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT hominlee datacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT byungguanyoon datacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT fayoungso datacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry AT jongpiljeong datacentricapproachtodesignandanalysisofasurfaceinspectionsystembasedondeeplearningintheplasticinjectionmoldingindustry
_version_	1718410663640956928

A Data-Centric Approach to Design and Analysis of a Surface-Inspection System Based on Deep Learning in the Plastic Injection Molding Industry

Ejemplares similares