Reference-Free Near-Field Microwave Imaging Through Optimal Standoff Distance Control for High-Contrast Subsurface Anomaly Detection

Near-field microwave imaging is a promising non-contact method for detecting subsurface anomalies. However, the impact of the air gap between the antenna and target on the image quality has not been systematically studied. In addition, many reconstruction techniques require background data, making reference-free imaging important for practical applications. This study investigates how antenna-to-target spacing affects imaging using a 1-port reflection-based setup with four patch antennas operating at 1.5, 2.43, 3, and 4 GHz. A distinct “null-point”- the standoff distance where the reflection coefficient is minimized - is identified and validated as the optimal spacing via simulations and experiments. ABS, clay, and water were used as dielectric targets with a water-filled container embedded in clay to simulate anomalies. For solid ABS, the first null-point ranges from 0.125λ0at 1.5 GHz to 0.03λ0at 4 GHz, with a larger range for the water-filled ABS block. The simulation and experimental results were in close agreement. A novel multi-spacing reference-free approach to near-field microwave imaging for subsurface anomaly detection with high contrast was presented. At optimal spacing, the Michelson contrast for water anomaly detection at 3 GHz increased to 74.5% compared to 42.4% at non-optimal spacing. The final reference-free images produced using the multi-spacing algorithm provided a contrast of 93.6% in the simulation and 91.2% in the measurement, validating the superior contrast performance of this approach. The Dice Loss also decreased from 0.251 to 0.116, demonstrating improved localization accuracy. This approach offers a cost-effective and practical solution for real-world deployment, without requiring reference data.