Abstract:The rough surface of targets can cause complex impact on the amplitude, phase, polarization of the scattered fields at THz frequencies. For weighting and describing surface roughness effects, a Perfect Electric Conductor(PEC) cone is taken as an example to illustrate the characteristics of the scattered fields of rough surface in radar imaging. The model of rough cone surface is obtained based on the spectral method. Multi-angle, multi-frequency, full polarized monostatic scattered fields are obtained by utilizing the high frequency methods of computational electromagnetics. Then the targets images are obtained by using the duplicate integral method according to the rotation model. Obvious differences between rough and ideal smooth cone can be found from the reconstruction images. The simulation results show that as the roughness degree increases, the energy distribution of rough cone images departs from those highlighting regions in smooth cone images; and the cross-polarized images have better potential of describing the target profile comparing to the like-polarized images.