Non-destructive Crack Detection and Classification for Waterproof-coated Concrete Surfaces by Millimeter-wave Imaging

被引：0

作者：

Hirata, Akihiko ^{[1
]}

Nakasizuka, Makoto ^{[1
]}

Sudo, Keiichi ^{[2
]}

机构：

[1] Chiba Inst Technol, Narashino, Chiba, Japan

[2] Ais Engn, Tokyo, Japan

来源：

2018 15TH EUROPEAN RADAR CONFERENCE (EURAD) | 2018年

关键词：

millimeter wave radar; machine learning; propagation;

D O I：

暂无

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

This paper presents non-destructive millimeterwave (MMW) imaging of sub-millimeter wide cracks on concrete surface that are covered by waterproof coating. Measurement of near-field scattering at cracks enables the detection of 0.3-mm-wide cracks by using 76.5-GHz MMW signal whose wavelength is about 3.9 mm. However, the roughness of the waterproof coating also causes near-field scattering, which makes it difficult to identify the cracks and surface roughness of waterproof coating. In order to increase the accuracy of crack detection, we simulated the reflected MMW signals at concrete cracks and surface roughness of waterproof coating, and applied a learning based classifier to these simulation results in order to classify concrete cracks and surface roughness of waterproof coating in the MMW images, and the classification accuracy of 97.5 % was achieved.

引用

页码：265 / 268

页数：4