Image-Based Machine Learning of In Situ Captured Images of Die-Workpiece Interface in Forging

Matsumoto, Ryo; Kunisawa, Shusuke; Nakamura, Yusaku; Utsunomiya, Hiroshi

doi:https://doi.org/10.4028/p-ex7ohs

Image-Based Machine Learning of In Situ Captured Images of Die-Workpiece Interface in Forging

Matsumoto, Ryo; Kunisawa, Shusuke; Nakamura, Yusaku et al.

Defect and Diffusion Forum, 2025, 441(9), 39-43

Number of Access:493（2025-08-14 18:26 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/11094/100688

Link to primary information

File	Format	Terms of use	Size	Views	Date.Available	Description	information
DDF_441_9_39	pdf	None	306 KB	24	2025.03.28

Item Information

Output File Export EndNote Basic Export Mendeley

Title	Image-Based Machine Learning of In Situ Captured Images of Die-Workpiece Interface in Forging
Creator	NRID 1000050362645 1000050362645 ORCID 0000-0002-4477-2139 0000-0002-4477-2139 researchmap permalink ryo-matsumoto ryo-matsumoto Scopus Author ID 35512063000 35512063000 CreatorName Matsumoto, Ryo
	CreatorName Kunisawa, Shusuke
	CreatorName Nakamura, Yusaku
	NRID 1000080252584 1000080252584 researchmap permalink read0042799 read0042799 Scopus Author ID 7102139699 7102139699 CreatorName Utsunomiya, Hiroshi
Subject	Forging
	Extrusion
	Lubrication
	Image analysis
	Machine learning
Description	Matsumoto Ryo, Kunisawa Shusuke, Nakamura Yusaku, et al. Image-Based Machine Learning of In Situ Captured Images of Die-Workpiece Interface in Forging. Defect and Diffusion Forum 441, 39 (2025); https://doi.org/10.4028/p-ex7ohs.
Abstract	The lubricant thickness in cold forging was estimated by machine learning of the in situ captured images of the die–workpiece contact interface. The images were in situ captured by a high-speed camera from the backside of the transparent glass die during forging of commercially pure aluminum workpiece. On the other hand, the images of the lubricated workpiece were individually captured as training images for random forest with classification. The classification accuracy of the lubricant thickness was confirmed to be approximately 75% (classification ability: 5–10 μm in lubricant thickness) in the training images with 22,500 px (50 px/mm). The in situ captured images of the die–workpiece contact interface during forging were classified by random forest using the training images. The estimated lubricant thickness of the in situ captured image almost agreed with the lubricant thickness estimated from the mean brightness value of the in situ captured image.
Conference Information	Conference Name the International Conference on Tribology in Manufacturing Processes（ICTMP） Conference Sequence 10 Conference Date June 26th to 28th, 2024 Conference Venue Universitat Politècnica de València Conference Place Alcoy Conference Country Spain
Publisher	Trans Tech Publications, Ltd.
Source Title	Defect and Diffusion Forum
Volume (Issue)	441(9)
Page	39-43
Date of Issued	2025-03-26
Language	English
Handle URL	https://hdl.handle.net/11094/100688
eISSN	16629507
Relation.isVersionOf	DOI (Publisher Version) https://doi.org/10.4028/p-ex7ohs
Access Rights	open access
oaire:version	Author’s Original
Category	会議発表論文 Conference Paper

Resource Type	conference paper
Local Type	Conference Paper
DCTERMS.bibliographicCitation	Defect and Diffusion Forum. 2025, 441(9), p. 39-43
DC.title	Image-Based Machine Learning of In Situ Captured Images of Die-Workpiece Interface in Forging
DC.creator	Matsumoto, Ryo
	Kunisawa, Shusuke
	Nakamura, Yusaku
	Utsunomiya, Hiroshi
DC.publisher	Trans Tech Publications, Ltd.
DC.language' scheme='DCTERMS.RFC1766	English
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2025-03-26
DC.identifier	https://doi.org/10.4028/p-ex7ohs
DC.subject	Forging
	Extrusion
	Lubrication
	Image analysis
	Machine learning
DC.description	Matsumoto Ryo, Kunisawa Shusuke, Nakamura Yusaku, et al. Image-Based Machine Learning of In Situ Captured Images of Die-Workpiece Interface in Forging. Defect and Diffusion Forum 441, 39 (2025); https://doi.org/10.4028/p-ex7ohs.
DCTERMS.abstract	The lubricant thickness in cold forging was estimated by machine learning of the in situ captured images of the die–workpiece contact interface. The images were in situ captured by a high-speed camera from the backside of the transparent glass die during forging of commercially pure aluminum workpiece. On the other hand, the images of the lubricated workpiece were individually captured as training images for random forest with classification. The classification accuracy of the lubricant thickness was confirmed to be approximately 75% (classification ability: 5–10 μm in lubricant thickness) in the training images with 22,500 px (50 px/mm). The in situ captured images of the die–workpiece contact interface during forging were classified by random forest using the training images. The estimated lubricant thickness of the in situ captured image almost agreed with the lubricant thickness estimated from the mean brightness value of the in situ captured image.
DC.format	100688
citation_title	Image-Based Machine Learning of In Situ Captured Images of Die-Workpiece Interface in Forging
citation_author	Matsumoto, Ryo
	Kunisawa, Shusuke
	Nakamura, Yusaku
	Utsunomiya, Hiroshi
citation_publisher	Trans Tech Publications, Ltd.
citation_language	English
citation_date	2025-03-26
citation_journal_title	Defect and Diffusion Forum
citation_volume	441
citation_issue	9
citation_firstpage	39
citation_lastpage	43
citation_doi	https://doi.org/10.4028/p-ex7ohs
citation_public_url	https://hdl.handle.net/11094/100688

Image-Based Machine Learning of In Situ Captured Images of Die-Workpiece Interface in Forging

Link to primary information

Item Information

Show details