Real-Time Super-Resolution Microwave Imaging Using Gamma Transformation

Microwave imaging is a promising medical imaging modality due to its low cost, safety, and portability, making it suitable for point-of-care applications. However, its clinical adoption is limited by low resolution and poor signal-to-clutter ratio (SCR), due to the heterogeneous and lossy nature of human tissues. While recent algorithms, such as various types of delay-multiply-and-sum (DMAS), improve image quality, they suffer from high computational complexity and relatively poor resolution, which limits their real-time clinical use. This paper proposes an imaging method based on Gamma transformation, inspired by the nonlinear response of human vision to brightness. The technique applies a power-law transformation to properly aligned spatial multi-static signals scattered from the imaged object. This approach enhances image contrast and resolution while maintaining linear computational complexity, thereby significantly reducing processing time. Theoretical analysis shows that a 2nd-order Gamma transformation achieves image quality comparable to DMAS but is considerably faster. Numerical and experimental results using head imaging systems with 32 and 16 antennas operating across the 0.75-1.8 GHz band confirm the method’s superior performance. A 16th-order Gamma transformation achieves a high SCR of 28.6 dB in numerical simulations and 4.79 dB in experimental studies, compared to less than 4 dB and 1 dB, respectively, for DMAS. The method also demonstrates super-resolution capability, resolving targets separated by 1 mm compared to 3 mm for DMAS. Additionally, it offers significantly faster computation time (2 seconds compared to 27 minutes in DMAS), demonstrating its potential as a reliable medical imaging method.