This paper describes a suite of techniques for the autonomous detection of moving targets by processing electro-optical sensor imagery (such as visible or infrared imagery). A new algorithm is proposed for the solution of an important class of detection problems: the detection of small, barely discernible, moving objects in presence of correlated fluctuating background and constant background intensity. We assume that ratio of the correlated fluctuating background variance to constant background intensity is not known and the constant background intensity is more than its variance. A comparison of the performance of the well-known and proposed algorithms is included. Proposed algorithm shows improvement in performance over the well-known algorithm in reducing of the constant background intensity and it is suboptimal in presence of correlated fluctuating background.
Golikov, V. & Lebedeva, O. (2006). Suboptimal Detection of 3-dimensional Optical Moving Target. Científica: La Revista Mexicana de Ingeniería Electromecánica, 10(3), pp. 107-111.