Landmines and improvised explosive devices are a major source of casualties in both current and former conflict zones. These hidden killers remain in the ground for years after conflict has ceased, killing and maiming indiscriminately. Although estimates vary, agencies including the Red Cross and the United Nations believe that there are between 60 and 70 million active landmines in the ground buried across 70 countries around the globe. Every year approximately 26,000 people are maimed or killed by landmines and 8,000 to 10,000 of these victims are children. Although landmines cost as little as $3 to produce, their presence inflicts a tremendous cost, especially in developing areas.
Our research for landmine and improvised explosive device remediation focuses on improved detection through a combination of physics-based signal processing, machine learning, and computer vision. Our current research focuses on techniques for context-dependent classification, and the incorporation of techniques from the computer vision literature into the field of landmine detection. We also make use of cutting-edge Bayesian statistical models to ground our research in a probabilistic framework.
Our research has resulted in algorithms currently deployed in real-life landmine remediation operations overseas.