Browse by author
Lookup NU author(s): Mohammed Buhari, Professor Gui Yun TianORCiD, Dr Rajesh TiwariORCiD
This is the authors' accepted manuscript of an article that has been published in its final definitive form by IEEE, 2019.
For re-use rights please refer to the publisher's terms and conditions.
IEEE In non-destructive testing and evaluation, microwave-based synthetic aperture radar (SAR) imaging have shown great potential in the detection of defects on buried objects such as pipes. However, due to pipe curvature and high standoff distance when inspecting an insulated pipe, the useful defect information used to characterise the pipe image is lost as a result of low signal-to-noise ratio (SNR) resulting in a blurred and unfocused image. In this paper, we proposed a robust microwave-based SAR imaging using autofocus range-Doppler algorithm (RDA) for the inspection of an insulated pipe. Singular value decomposition (SVD) is used to mitigate the effect of the insulation layer by removing dominant singular values that characterise the insulation layer, and the autofocus RDA is designed to refocus the SAR image using RDA residual refocusing. SNR, improvement factor (IF) and squared error (SE) are used to evaluate the qualitative image information of the defect on the pipe. Experimental results showed the efficacy of the method in detecting defects on an insulated pipe, in particular, a significant reduction in the noise content of the image compared to the known SAR Omega-k algorithm. It was found that the autofocus RDA gave higher values of SNR and IF (3 dB and 6 dB) compared to the Omega-k algorithm (-1 dB and 2 dB) respectively.
Author(s): Buhari MD, Tian GY, Tiwari R
Publication type: Article
Publication status: Published
Journal: IEEE Sensors Journal
Year: 2019
Volume: 19
Issue: 5
Pages: 1777-1787
Print publication date: 01/03/2019
Online publication date: 02/11/2018
Acceptance date: 02/04/2018
Date deposited: 19/11/2018
ISSN (print): 1530-437X
ISSN (electronic): 1558-1748
Publisher: IEEE
URL: https://doi.org/10.1109/JSEN.2018.2879348
DOI: 10.1109/JSEN.2018.2879348
Altmetrics provided by Altmetric
