A Novel THz Differential Spectral Clustering Recognition Method Based on t-SNE

We apply time-domain spectroscopy (THz) imaging technology to perform nondestructive detection on three industrial ceramic matrix composite (CMC) samples and one silicon slice with defects. In terms of spectrum recognition, a low-resolution THz spectrum image results in an ineffective recognition on...

Full description

Saved in:
Bibliographic Details
Main Authors: Tie-Jun Li, Chih-Cheng Chen, Jian-jun Liu, Gui-fang Shao, Christopher Chun Ki Chan
Format: Article
Language:English
Published: Wiley 2020-01-01
Series:Discrete Dynamics in Nature and Society
Online Access:http://dx.doi.org/10.1155/2020/6787608
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850163878294454272
author Tie-Jun Li
Chih-Cheng Chen
Jian-jun Liu
Gui-fang Shao
Christopher Chun Ki Chan
author_facet Tie-Jun Li
Chih-Cheng Chen
Jian-jun Liu
Gui-fang Shao
Christopher Chun Ki Chan
author_sort Tie-Jun Li
collection DOAJ
description We apply time-domain spectroscopy (THz) imaging technology to perform nondestructive detection on three industrial ceramic matrix composite (CMC) samples and one silicon slice with defects. In terms of spectrum recognition, a low-resolution THz spectrum image results in an ineffective recognition on sample defect features. Therefore, in this article, we propose a spectrum clustering recognition model based on t-distribution stochastic neighborhood embedding (t-SNE) to address this ineffective sample defect recognition. Firstly, we propose a model to recognize a reduced dimensional clustering of different spectrums drawn from the imaging spectrum data sets, in order to judge whether a sample includes a feature indicating a defect or not in a low-dimensional space. Second, we improve computation efficiency by mapping spectrum data samples from high-dimensional space to low-dimensional space by the use of a manifold learning algorithm (t-SNE). Finally, to achieve a visible observation of sample features in low-dimensional space, we use a conditional probability distribution to measure the distance invariant similarity. Comparative experiments indicate that our model can judge the existence of sample defect features or not through spectrum clustering, as a predetection process for image analysis.
format Article
id doaj-art-e9770e962f724e89a92c28dfae8acc4f
institution OA Journals
issn 1026-0226
1607-887X
language English
publishDate 2020-01-01
publisher Wiley
record_format Article
series Discrete Dynamics in Nature and Society
spelling doaj-art-e9770e962f724e89a92c28dfae8acc4f2025-08-20T02:22:06ZengWileyDiscrete Dynamics in Nature and Society1026-02261607-887X2020-01-01202010.1155/2020/67876086787608A Novel THz Differential Spectral Clustering Recognition Method Based on t-SNETie-Jun Li0Chih-Cheng Chen1Jian-jun Liu2Gui-fang Shao3Christopher Chun Ki Chan4College of Information Engineering, Jimei University, Fujian Province, Xiamen 361021, ChinaCollege of Information Engineering, Jimei University, Fujian Province, Xiamen 361021, ChinaShaoguan University, Guangzhou Province, Shaoguan 512005, ChinaDepartment of Automation, Xiamen University, Fujian Province, Xiamen 361005, ChinaDepartment of Information Management, Chaoyang University of Technology, Taichung, TaiwanWe apply time-domain spectroscopy (THz) imaging technology to perform nondestructive detection on three industrial ceramic matrix composite (CMC) samples and one silicon slice with defects. In terms of spectrum recognition, a low-resolution THz spectrum image results in an ineffective recognition on sample defect features. Therefore, in this article, we propose a spectrum clustering recognition model based on t-distribution stochastic neighborhood embedding (t-SNE) to address this ineffective sample defect recognition. Firstly, we propose a model to recognize a reduced dimensional clustering of different spectrums drawn from the imaging spectrum data sets, in order to judge whether a sample includes a feature indicating a defect or not in a low-dimensional space. Second, we improve computation efficiency by mapping spectrum data samples from high-dimensional space to low-dimensional space by the use of a manifold learning algorithm (t-SNE). Finally, to achieve a visible observation of sample features in low-dimensional space, we use a conditional probability distribution to measure the distance invariant similarity. Comparative experiments indicate that our model can judge the existence of sample defect features or not through spectrum clustering, as a predetection process for image analysis.http://dx.doi.org/10.1155/2020/6787608
spellingShingle Tie-Jun Li
Chih-Cheng Chen
Jian-jun Liu
Gui-fang Shao
Christopher Chun Ki Chan
A Novel THz Differential Spectral Clustering Recognition Method Based on t-SNE
Discrete Dynamics in Nature and Society
title A Novel THz Differential Spectral Clustering Recognition Method Based on t-SNE
title_full A Novel THz Differential Spectral Clustering Recognition Method Based on t-SNE
title_fullStr A Novel THz Differential Spectral Clustering Recognition Method Based on t-SNE
title_full_unstemmed A Novel THz Differential Spectral Clustering Recognition Method Based on t-SNE
title_short A Novel THz Differential Spectral Clustering Recognition Method Based on t-SNE
title_sort novel thz differential spectral clustering recognition method based on t sne
url http://dx.doi.org/10.1155/2020/6787608
work_keys_str_mv AT tiejunli anovelthzdifferentialspectralclusteringrecognitionmethodbasedontsne
AT chihchengchen anovelthzdifferentialspectralclusteringrecognitionmethodbasedontsne
AT jianjunliu anovelthzdifferentialspectralclusteringrecognitionmethodbasedontsne
AT guifangshao anovelthzdifferentialspectralclusteringrecognitionmethodbasedontsne
AT christopherchunkichan anovelthzdifferentialspectralclusteringrecognitionmethodbasedontsne
AT tiejunli novelthzdifferentialspectralclusteringrecognitionmethodbasedontsne
AT chihchengchen novelthzdifferentialspectralclusteringrecognitionmethodbasedontsne
AT jianjunliu novelthzdifferentialspectralclusteringrecognitionmethodbasedontsne
AT guifangshao novelthzdifferentialspectralclusteringrecognitionmethodbasedontsne
AT christopherchunkichan novelthzdifferentialspectralclusteringrecognitionmethodbasedontsne