Abstract: Self target aiming and pre-processing of visual data is one of the biggest challenges in solving red-world problem using autonomous robots. Conventional methods generally depend on CCD cameras and computers associated with base station which doesn’t implies real world interaction. Our approach attempts to solve this problem by using computational sensors and small/inexpensive embedded processors. The computational sensors are custom designed to reduce the amount of data collected, to extract only relevant information and to present this information to the microcontroller , in a format which minimizes post-processing latency. Consequently, the post-processors are required to perform only high level computation and decision making on given data. The computational sensors, however, have wide applications in many problems that require image pre-processing such as edge detection, motion detection, centroid localization and other spatiotemporal processing.

Keywords: Computational sensor, Pre-processing, Motion centroid computation, Weapon identification.