PointNet + + Based Concealed Object Classification Utilizing an FMCW Millimeter-Wave Radar

Wang, Yaheng; Su, Jie; Murakami, Hironaru et al.

Journal of Infrared, Millimeter, and Terahertz Waves, 2024, 45(11-12), 1040-1057

Number of Access:478（2025-09-09 06:48 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/11094/98397

Link to primary information 

File Format Terms of use Size Views Date.Available Description information
JInfMilTeraWaves_s10762-024-01017-5 pdf None 3.97 MB 452 2024.11.08  
Creative Commons : Attribution

Item Information

Output File Export EndNote Basic Export Mendeley

Title
PointNet + + Based Concealed Object Classification Utilizing an FMCW Millimeter-Wave Radar
Creator
CreatorName
Wang, Yaheng
CreatorName
Su, Jie
NRID
1000030219901 1000030219901
ORCID
0000-0003-0399-8104 0000-0003-0399-8104
researchmap permalink
hironaru35020227 hironaru35020227
Scopus Author ID
35412853700 35412853700
CreatorName
Murakami, Hironaru
NRID
1000040207593 1000040207593
ORCID
0000-0002-9284-3501 0000-0002-9284-3501
researchmap permalink
read0051752 read0051752
Scopus Author ID
7006497099 7006497099
CreatorName
Tonouchi, Masayoshi
Subject
Deep learning
Millimeter wave imaging
Millimeter wave radar
Object classification
Description
The version of record of this article, first published in Journal of Infrared, Millimeter, and Terahertz Waves, is available online at Publisher’s website: https://doi.org/10.1007/s10762-024-01017-5.
Abstract
In the field of millimeter-wave (MMW) imaging, the integration of artificial intelligence (AI) has emerged as a crucial solution for addressing automation challenges. In this study, concealed object classification was successfully achieved on point cloud data from MMW radar high-precision imaging using the PointNet + + deep learning method. The utilized dataset comprises point cloud data generated through the transformation of 3D models and reconstruction of physical objects with an accuracy of less than 1 mm via MMW radar scanning. Classification accuracy was significantly improved by introducing data enhancement techniques, including the generation of homologous data and optimization of sampling points. After several evaluations, 300 epochs of training were conducted using 8192 sampling points, the results showed an accuracy of 0.998 for the training dataset and 0.996 for the test dataset. Moreover, evaluations of samples not included in the original dataset as well as multi-surface scans of concealed objects within the cardboard both resulted in correct predictions, which further validates the effectiveness and reliability of the study and demonstrates the potential of AI applied to MMW imaging.
Publisher
Springer
Source Title
Journal of Infrared, Millimeter, and Terahertz Waves
Volume (Issue)
45(11-12)
Page
1040-1057
Date of Issued
2024-10-11
Language
English
Handle URL
https://hdl.handle.net/11094/98397
PISSN
18666892
eISSN
18666906
Relation.isIdenticalTo
DOI (Publisher Version)
https://doi.org/10.1007/s10762-024-01017-5
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