Infrared Detection of Buried Land Mines Based on Texture Modeling

A parametric model for the infrared signature caused by a buried land mine is presented. Further, two ways ofmodeling the colored background noise, is proposed. In the first, it is assumed the noise can be approximated by anautoregressive process, while in the second, the statistics of the noise is described using recent development in texturemodeling, the so called FRAME method. Given an a priori distribution of the mine parameters in combination with atrained noise distribution, a Bayesian detector is derived. Experiments indicate that significant gains in performancecan be achieved as compared to the standard detector used, which correlates the infrared image with the knownmine shape and thresholds the square of the output