A Sensor with Biological Preprocessing Features

This report discusses on A Sensor with Biological Preprocessing Features during the period February 1, 1986 to October 31, 1987. The first part of the report reviews the basic geometric sensor configuration and its mapping to computation plane. Three basic configurations are discussed and their application to edge and short range motion detection follows. The next section covers a qualitative long range motion detection algorithm. The algorithm was tested first with synthetic images and next real images, and in both cases the results were very satisfactory. The algorithm is now being tested with multiple images. A one line correlation tracker is discussed in the next section. The correlation tracker can be implemented with parallel architecture and is, very fast. Tests have been performed with satisfactory results using real images. Pattern recognition using BVS together with a class transforms which are invariant to translations is discussed. It should be possible to implement this pattern recognition system using neural networks. A preliminary neural network implementation is given. The use of learning neural nets will allow recognition of sensor follows. A hardware configuration is proposed to map a rectangular array into a log-spiral array; and an error analysis is performed. Finally, depth computation from optical flow using the new sensor is discussed. Keywords: Cybernetics; Image processing; Spatial distribution; Two dimensional; Three dimensional; Computer simulation.