{"title":"Unsupervised wood species identification based on multiobjective optimal clustering and feature fusion","authors":"Yuan Wang, Wen-Jin Ma, Meng Yang, Ren-He Qu, Stavros Avramidis","doi":"10.1007/s00226-025-01636-8","DOIUrl":null,"url":null,"abstract":"<div><p>This paper proposes an unsupervised wood species identification approach utilizing multiobjective optimization clustering and feature fusion. To address the inherent limitations of single-band spectra in capturing comprehensively wood characteristics, this approach integrates preprocessed low-dimensional terahertz (THz) and hyperspectral data. Additionally, to address the challenge of selecting the optimal k-value in clustering, an unsupervised wood clustering algorithm was developed, employing multiobjective optimization and evolutionary algorithms. This proposed algorithm incorporated a prototype coding method for initialization, density peak clustering for pattern identification, and an improved firefly optimization algorithm to introduce cross-variation and maintain population diversity. To further refine the clustering process, a selection operator based on grid division and fast non-dominated sorting was designed, optimizing the clustering performance. The model was evaluated on a dataset containing hyperspectral and THz spectra from 400 samples, representing ten wood species—five coniferous and five broadleaf species. Experimental results indicated that fusing the spectral data resulted in a 3.5% increase in clustering purity compared to individual datasets. Moreover, the proposed algorithm outperformed established clustering methods such as DBSCAN, OPTICS, and density peak clustering, achieving a maximum clustering purity of 91.25% across both internal and external clustering metrics. These findings demonstrate the effectiveness of the multi-spectral fusion approach and the proposed algorithm in enhancing wood species identification accuracy, offering a promising avenue for improving non-destructive evaluation methods in forestry and material sciences.</p></div>","PeriodicalId":810,"journal":{"name":"Wood Science and Technology","volume":"59 2","pages":""},"PeriodicalIF":3.1000,"publicationDate":"2025-02-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Wood Science and Technology","FirstCategoryId":"88","ListUrlMain":"https://link.springer.com/article/10.1007/s00226-025-01636-8","RegionNum":2,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"FORESTRY","Score":null,"Total":0}
引用次数: 0
Abstract
This paper proposes an unsupervised wood species identification approach utilizing multiobjective optimization clustering and feature fusion. To address the inherent limitations of single-band spectra in capturing comprehensively wood characteristics, this approach integrates preprocessed low-dimensional terahertz (THz) and hyperspectral data. Additionally, to address the challenge of selecting the optimal k-value in clustering, an unsupervised wood clustering algorithm was developed, employing multiobjective optimization and evolutionary algorithms. This proposed algorithm incorporated a prototype coding method for initialization, density peak clustering for pattern identification, and an improved firefly optimization algorithm to introduce cross-variation and maintain population diversity. To further refine the clustering process, a selection operator based on grid division and fast non-dominated sorting was designed, optimizing the clustering performance. The model was evaluated on a dataset containing hyperspectral and THz spectra from 400 samples, representing ten wood species—five coniferous and five broadleaf species. Experimental results indicated that fusing the spectral data resulted in a 3.5% increase in clustering purity compared to individual datasets. Moreover, the proposed algorithm outperformed established clustering methods such as DBSCAN, OPTICS, and density peak clustering, achieving a maximum clustering purity of 91.25% across both internal and external clustering metrics. These findings demonstrate the effectiveness of the multi-spectral fusion approach and the proposed algorithm in enhancing wood species identification accuracy, offering a promising avenue for improving non-destructive evaluation methods in forestry and material sciences.
期刊介绍:
Wood Science and Technology publishes original scientific research results and review papers covering the entire field of wood material science, wood components and wood based products. Subjects are wood biology and wood quality, wood physics and physical technologies, wood chemistry and chemical technologies. Latest advances in areas such as cell wall and wood formation; structural and chemical composition of wood and wood composites and their property relations; physical, mechanical and chemical characterization and relevant methodological developments, and microbiological degradation of wood and wood based products are reported. Topics related to wood technology include machining, gluing, and finishing, composite technology, wood modification, wood mechanics, creep and rheology, and the conversion of wood into pulp and biorefinery products.