Insights into a new overclustering technique using machine learning for a self-selecting bin-restricted colour sorting setup for light red meranti (Rubroshorea spp.)

Tan, Chiat Oon; Ogata, Shigenobu; Yap, Hwa Jen; Nakamoto, Ichiro; Usop, Zuriani; Fauthan, Mohd ’Akashah; Liew, Shaer Jin; Ng, Siew Cheok

doi:https://doi.org/10.1007/s00107-025-02280-6

Insights into a new overclustering technique using machine learning for a self-selecting bin-restricted colour sorting setup for light red meranti (Rubroshorea spp.)

Tan, Chiat Oon; Ogata, Shigenobu; Yap, Hwa Jen et al.

European Journal of Wood and Wood Products, 2025, 83(4), 131

Number of Access:31（2025-11-07 13:01 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/11094/102590

Link to primary information

File	Format	Terms of use	Size	Views	Date.Available	Description	information
EurJWoodWoodProd_83_4	pdf	None	3.54 MB	14	2025.08.29

Creative Commons : Attribution

Item Information

Output File Export EndNote Basic Export Mendeley

Title	Insights into a new overclustering technique using machine learning for a self-selecting bin-restricted colour sorting setup for light red meranti (Rubroshorea spp.)
Creator	CreatorName Tan, Chiat Oon
	NRID 1000020273584 1000020273584 ORCID 0000-0002-9072-4496 0000-0002-9072-4496 researchmap permalink read0185246 read0185246 Scopus Author ID 7201796489 7201796489 CreatorName Ogata, Shigenobu
	CreatorName Yap, Hwa Jen
	CreatorName Nakamoto, Ichiro
	CreatorName Usop, Zuriani
	CreatorName Fauthan, Mohd ’Akashah
	CreatorName Liew, Shaer Jin
	CreatorName Ng, Siew Cheok
Description	The version of record of this article, first published in European Journal of Wood and Wood Products, is available online at Publisher’s website: https://doi.org/10.1007/s00107-025-02280-6.
Abstract	Timber colour sorting is an important woodworking process in producing a homogeneously coloured and pleasant looking product. However, for multispecific timber such as light red meranti, LRM (Rubroshorea spp.), which spans a wide gamut of colours, there is an antagonistic compromise between having good separability of colour and the number of bins. This research attempts to solve this by intentionally overclustering the intensity gamut and then automating the selection of ideal colour sorting bins (CSB) for a given batch size to produce high-similarity coloured sorting. 178,327 unique LRM wood samples collected over 8 months of production were used. Machine learning clustering algorithms such as k-means and Otsu multithresholding were tested against percentile and equal spacing methods. Batch sizes of 250 (B250) and 1,000 (B1000) pieces were evaluated. Maximum likelihood estimation was tested against statistical methods to select the CSB, and ideal overcluster setups were determined using the average delta E (ΔE*₀₀) assessment. The ‘burn-in rates’ of 3–30 pieces were then evaluated. For the B250 four-bin setup, six overclusters (6C4) performed best, with a recommended ‘burn-in rate’ of 12 pieces. For B1000, 5C4 performed best with a ‘burn-in rate’ of 10 pieces. The 4C3 configuration and the ‘burn-in rate’ of 10 pieces were found to be the best for three-CSB for both B250 and B1000. This study shows the feasibility of using machine learning to automate the bin selection process when the overclustering technique is used to improve colour sorting in situations with a restricted number of bins.
Publisher	Springer Science and Business Media Deutschland GmbH
Source Title	European Journal of Wood and Wood Products
Volume (Issue)	83(4)
Page	131
Date of Issued	2025-06-25
Language	English
Handle URL	https://hdl.handle.net/11094/102590
PISSN	00183768
eISSN	1436736X
Relation.isIdenticalTo	DOI (Publisher Version) https://doi.org/10.1007/s00107-025-02280-6
Access Rights	open access
CopyRight	This article is licensed under a Creative Commons Attribution 4.0 International License.
oaire:version	Version of Record
Category	学術雑誌論文 Journal Article

Resource Type	journal article
Local Type	Journal Article
DCMI Type	Text
DCTERMS.bibliographicCitation	European Journal of Wood and Wood Products. 2025, 83(4), p. 131
DC.title	Insights into a new overclustering technique using machine learning for a self-selecting bin-restricted colour sorting setup for light red meranti (Rubroshorea spp.)
DC.creator	Tan, Chiat Oon
	Ogata, Shigenobu
	Yap, Hwa Jen
	Nakamoto, Ichiro
	Usop, Zuriani
	Fauthan, Mohd ’Akashah
	Liew, Shaer Jin
	Ng, Siew Cheok
DC.publisher	Springer Science and Business Media Deutschland GmbH
DC.language' scheme='DCTERMS.RFC1766	English
DC.type' scheme='DCTERMS.DCMIType	Text
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2025-06-25
DC.identifier	https://doi.org/10.1007/s00107-025-02280-6
DC.description	The version of record of this article, first published in European Journal of Wood and Wood Products, is available online at Publisher’s website: https://doi.org/10.1007/s00107-025-02280-6.
DCTERMS.abstract	Timber colour sorting is an important woodworking process in producing a homogeneously coloured and pleasant looking product. However, for multispecific timber such as light red meranti, LRM (Rubroshorea spp.), which spans a wide gamut of colours, there is an antagonistic compromise between having good separability of colour and the number of bins. This research attempts to solve this by intentionally overclustering the intensity gamut and then automating the selection of ideal colour sorting bins (CSB) for a given batch size to produce high-similarity coloured sorting. 178,327 unique LRM wood samples collected over 8 months of production were used. Machine learning clustering algorithms such as k-means and Otsu multithresholding were tested against percentile and equal spacing methods. Batch sizes of 250 (B250) and 1,000 (B1000) pieces were evaluated. Maximum likelihood estimation was tested against statistical methods to select the CSB, and ideal overcluster setups were determined using the average delta E (ΔE*₀₀) assessment. The ‘burn-in rates’ of 3–30 pieces were then evaluated. For the B250 four-bin setup, six overclusters (6C4) performed best, with a recommended ‘burn-in rate’ of 12 pieces. For B1000, 5C4 performed best with a ‘burn-in rate’ of 10 pieces. The 4C3 configuration and the ‘burn-in rate’ of 10 pieces were found to be the best for three-CSB for both B250 and B1000. This study shows the feasibility of using machine learning to automate the bin selection process when the overclustering technique is used to improve colour sorting in situations with a restricted number of bins.
DC.rights	This article is licensed under a Creative Commons Attribution 4.0 International License.
DC.format	102590
citation_title	Insights into a new overclustering technique using machine learning for a self-selecting bin-restricted colour sorting setup for light red meranti (Rubroshorea spp.)
citation_author	Tan, Chiat Oon
	Ogata, Shigenobu
	Yap, Hwa Jen
	Nakamoto, Ichiro
	Usop, Zuriani
	Fauthan, Mohd ’Akashah
	Liew, Shaer Jin
	Ng, Siew Cheok
citation_publisher	Springer Science and Business Media Deutschland GmbH
citation_language	English
citation_date	2025-06-25
citation_journal_title	European Journal of Wood and Wood Products
citation_volume	83
citation_issue	4
citation_firstpage	131
citation_doi	https://doi.org/10.1007/s00107-025-02280-6
citation_public_url	https://hdl.handle.net/11094/102590

Insights into a new overclustering technique using machine learning for a self-selecting bin-restricted colour sorting setup for light red meranti (Rubroshorea spp.)

Link to primary information

Item Information

Show details