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

Tie-Jun Li,Chih-Cheng Chen,Gui-Fang Shao,Christopher Chun Ki Chan,Jian-Jun Liu

doi:10.1155/2020/6787608

Abstract

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.

Highlights

Nondestructive testing is one of the most significant applications of terahertz technology, and terahertz time-domain spectroscopy (THz—TDS) system is a commonly used technique [1, 2]
In order to improve the efficiency of t-distribution stochastic neighborhood embedding (t-stochastic neighbor embedding (SNE)), the principal components analysis (PCA) method is usually introduced first to reduce the dimension of a high-dimensional sample point data set to 50 dimensions, and t-SNE is used for cluster recognition. e specific pseudocode is shown in Algorithm 1
The data set of abnormal spectrum is significantly smaller than that of normal structure samples and is in free state, and the results are in line with the predicted analysis

Summary

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

In terms of spectrum recognition, a low-resolution THz spectrum image results in an ineffective recognition on sample defect features. Erefore, 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. 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 lowdimensional space. 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). 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

Introduction

Experiment and Analysis

Conclusion