Practical landmine detector using forward-looking ground penetrating radar

A practical landmine detector using forward-looking ground penetrating radar is presented. The detector uses a wavelet packet transform and the sequential feature selection algorithm to extract the most discriminant information distributed in the joint time-frequency domain for detecting landmines. A cascade training method that allows the detector to continue learning from errors made on unseen environments to improve its detection performance is also proposed. The testing results based on a recently conducted blind test are presented.