Robust IED and Landmine Detection Systems Using IR Images

Traditional landmine detection techniques are both dangerous and time consuming. Landmines can be square, round, cylindrical, or bar shaped. The casing can be metal, plastic, or wood. These characteristics make landmine detection challenging. In Phase I, we have developed new methods that improve the performance of both surface and buried mine detection. Extensive tests using actual MWIR images have shown extremely promising results. In Phase II, we will conduct further tests using additional ASTAMIDS data sets. Algorithms developed in Phase I will be further improved, which include an innovative image enhancement algorithm, new buried mine classifiers, better contour extraction method, and LMI based soft fusion. The improved system will be further adapted for IED detection through discriminating buried command wires using IR images collected at the field. We will also develop a commercial product for natural gas pipeline leak detection. Our Phase II prototype system is a combination of both software and hardware with various I/O interfaces to imaging sensors and existing ASTAMIDS systems. All algorithms will be implemented in C/C++ and a sophisticated graphical user interface will be developed in Microsoft VB.NET. Our system can be deployed at either aerial or ground platforms with near real time detection capabilities.