Performance Comparison of FBP and CS Methods on Microwave Tomography

Abstract

This paper presents the investigation and comparative analysis of image reconstruction performance on microwave tomography, namely filtered back projection (FBP) and compressive sensing (CS). The investigation is carried out by processing a picture of tree trunk object with a pixel size of 128 × 128. Performance comparison are assessed using mean square error (MSE) and structural similarity index measure (SSIM) parameters. The result of image reconstruction using CS method has better MSE and SSIM with the value of 0.001 and 0.837, respectively, for smaller image pixel, while the result of using FBP method has SSIM value of 0.901.