Abstract
This paper introduces a new rapid detection model with multi-row crossed antenna pairs and a corresponding algorithm for online rapid detection. The way of multi-row crossed antennas leads to data offset of adjacent columns, so a method of magnitude and phase restoration is discussed in the case of material movement. For reducing the effective testing scope and improving the resolution, a modified Vivaldi antenna with more concentrated E-field distribution was designed and fabricated, which operates at a ultra-wideband range from 3.9 to 40 GHz with S11 less than –10 dB for covering the detection frequency. To measure the dimension and accurately locate the positions of material defects, a proposed algorithm that combines 2-D cubic spline interpolation together with Laplace of Gaussian is introduced. This algorithm can extract the dimension of defects (5 mm minimum) from the original magnitudes and phases of materials. To evaluate the proposed method, the cystosepiment with a high-loss dielectric piece (5, 10, and 15 mm in diameter) inside was fabricated and an experimental testing system was established to achieve the rapid detection. Compared with the existed method with a-row antennas to detect, the proposed detection model of the multi-row antennas and corresponding algorithm are effective to locate the defect with a offset less than 0.5 cm.
Acknowledgements
The authors acknowledge the assistances and support from associate professor Guo at UESTC for paper modification.
Notes
No potential conflict of interest was reported by the authors.