2022
DOI: 10.1109/access.2022.3200473
|View full text |Cite
|
Sign up to set email alerts
|

Material Classification for Terahertz Images Based on Neural Networks

Abstract: Terahertz time-domain spectroscopy enables the extraction of electrical properties from materials. An extraction of the complex permittivity can be carried out with measurements in transmission or reflection geometry enabling the identification of materials. To perform an exact identification, the sample thickness and the angle of incidence of the terahertz radiation must be known. However, when those parameters are unknown and additionally the materials show strong absorbances, a precise differentiation betwe… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
9
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
7
1

Relationship

3
5

Authors

Journals

citations
Cited by 11 publications
(9 citation statements)
references
References 25 publications
0
9
0
Order By: Relevance
“…Furthermore, the method can be combined with terahertz ellipsometry to determine reference-free material parameters from the images [27]. Additionally, the use of neural networks could enable a classification of the measured material [28].…”
Section: Discussionmentioning
confidence: 99%
“…Furthermore, the method can be combined with terahertz ellipsometry to determine reference-free material parameters from the images [27]. Additionally, the use of neural networks could enable a classification of the measured material [28].…”
Section: Discussionmentioning
confidence: 99%
“…Examples of these databases are high-resolution transmission molecular absorption database (HITRAN) [57] and Cologne database for molecular spectroscopy (CDMS) [58]. In the future, these databases will play an important role in the development of new devices and can also be used for material identification by using machine learning [59]. Especially for strongly absorbing samples, systems with a high SNR are needed here.…”
Section: Comparision and Outlookmentioning
confidence: 99%
“…2. The visualization is for the material COC with the refractive index n = 1.52 (average of [20] and [21]) and the thicknesses d = 1 mm and d = 1.5 mm, respectively. The material losses are neglected.…”
Section: Single Fabry-pérot Resonatormentioning
confidence: 99%
“…Here, manufacturing is carried out by fused deposition modeling (FDM) 3D printing with the material COC. The low broadband absorption of COC, also known as TOPAS, in the terahertz region enables the manufacturing of high quality transmission devices like terahertz optics, waveguides, and filters [20][21][22][23]. Here, the 3D printer Ultimaker S5 is used with a 0.23 mm-diameter nozzle.…”
Section: Fabrication By 3d Printingmentioning
confidence: 99%