A framework for multiple target detection for surveillance

Detecting targets from an airborne platform is a difficult task due to associated platform disturbances. Target has to be distinguished from clutter. In order to achieve this objective, a connected component analysis (CCA) is utilized effectively for surveillance application in real time. CCA is performed on a binary image to distinctively identify multiple independent targets in a binary image. The various target detection methodologies along with verification techniques are applied on the gray level image. Several experimental results are also presented to support the validity of the method employed.