Applied modified nonlocal-means super-resolution method to microwave images

A microwave imaging device is used to catch the image from artificial phantom. This microwave imaging system is developed by Allwave corporation. One of the main techniques is called the field mapping algorithm (FMA). By using FMA, the electric field data of the scanning object can be obtained. After scanning, we use the simple threshold to eliminate unnecessary noise and uninterested signals. Then, the electric field data are normalized to generate pixel values. However, there are many drawbacks, such as low contrast, blurring, and lower spatial resolution in the microwave image. To overcome the lower spatial resolution drawback, we adopt a new image super-resolution (SR) reconstruction approach based on modified nonlocal-means (NLM) SR algorithm. Simulation results show that the proposed method successfully improves some perceptual quality of the super-resolved microwave images.