Antipersonnel mine detection on infrared images

Detection and clearance of buried mines is a big problem with lots of humanitarian, environmental and economic implications. Classic techniques, such as metal detectors, are not suitable for detecting mines with low or null metal content. In this work we propose a novel method based on the analysis of the different thermodynamic behaviour of the mines and the soil using a sequence of infrared images. In order to process this information, we propose the use of a specific type of neural network characterized by its parallel nature, high processing speed, low cost and power consumption and small size.