No abstract

No abstract

An opto-electronic associative neural network using a time division multiplexing technique (TDM) was first demonstrated. In the physically 32 neuron system, the eight state vectors could be stored with the division number 2. The experimental results verified the potential advantages and widespread applications of the TDM techniques. ABSTRACTWe describe a fornulation for 2-D neural nets which is suitabk for the processing of 2-D m y images. The formulation essentially processes a 2-D input array to be pre-and postmultiplied by a matrix of weights. Then the resulting 2-D array of data is processed via the usual sigmoidal thrcr holding and feedback which are characteristic of neural nets. We describe two electro-optical implementations of our formulation: (i) includes adaptations of the available incoherent optical techniques for the d -t i m e multiplication of 3 mamces, and (ii) includes adaptations of vector-mamx multiplication and the sum of the outer-product of vectors. The second requires reduced optical hardware for its implementation. AbstractMultitarget tracking over consecutive pairs of time frames is accomplished with a neural net. This involves position and velocity measurements of the targets and a quadratic neural energy function. Simulation data are presented, and an optical implementation is discussed.