GPR migration algorithm for landmines buried in inhomogeneous soil

Accident caused by antipersonnel (AP) landmines remaining from past conflicts have received considerable public exposure in the last several years. Detection of minimum-metal or plastic landmines by conventional metal detectors in a difficult and challenging task. Ground penetrating radar (GPR) can be an effective subsurface imaging tool for solution to landmine detection, because it can detect both the metal and nonmetal objects based on the dielectric properties. However, the detection of buried landmines by GPR normally suffers from very strong clutter that would decrease the signal-clutter ratio and deform the image of the landmine in GPR data. For enhancement of the signal-clutter ratio and improvement of the image quality, we developed a stepped-frequency continuous-wave array antenna ground penetrating radar system, called SAR-GPR. The system can acquire CMP data instead of generic common-offset data. Some data processing techniques have been in use to process CMP data, including velocity analysis, CMP stacking and migration. Also vertical variable velocity model is estimated and applied to normal moveout correction (NMO) and Kirchhoff migration