Fuzzy Automatic Detection of Landmines from Sensors Data

In this paper, two fuzzy algorithms for automatic decision making for antipersonnel landmine detection form sensors data. The first is a “feature in-decision out” fuzzy fusion algorithm for two sensors measurements, namely a ground penetrating radar (GPR) and a metal detector (MD). The inputs to the fuzzy fusion algorithm are features extracted from both GPR and MD measurements while the output is a decision if there is a land mine and at what depth it would be. Fuzzy fusion rules are extracted from training data through a fuzzy learning algorithm. The second is a 3D fuzzy template based automatic detection algorithm from a single sensor data, (GPR). A 3D template is chosen and another 3D fuzzy template is designed. The 3D fuzzy template, in which a data point is expressed as a trapezoidal fuzzy set, is extracted from experimental data. Landmine similarity for both the 3D template as well as the learnt fuzzy template is examined by a crisp similarity measure and a fuzzy similarity measure respectively. Results of both the two fuzzy decision making algorithms are presented which show their promises in landmine detection and its discrimination from other objects. Key-Words: Fuzzy decision making, landmine detection, sensor fusion