Multiparameter Adaptive Target Classification Using Full-Polarimetric GPR: A Novel Approach to Landmine Detection

Full-polarimetric ground penetrating radar (FP-GPR) can measure the ability of an object to change polarization electromagnetic waves. Compared traditional GPR, it has a stronger capability identify underground objects. In recent years, series decomposition methods have been applied FP-GPR data processing obtain polarimetric attributes and enhance targets identification. Different characterize different features target but there is still no effective way integrate these take their respective advantages for classification. this article, we propose particle center AdaBoost (PCAD) method achieve multiparameter adaptive The experimental results indicate that PCAD automatically select suitable parameters during training process targets. single-parameter classification based on average Bagging method, presents higher correct rates in Finally, proposed landmine detection. demonstrate composite scatterer generate surface scattering signals its dipole volume from interior; color-coded two-dimensional image by PCAD, distinguish landmines other