An Image Association Model of the Brodmann Areas

The ability to associate images is the basis for learning relationships involving vision, hearing, tactile sensation, and kinetic motion. A new architecture is described that has only local, recurrent connections, but can directly form global image associations. This architecture has many similarities to the structure of the neocortex, including the division into Brodmann areas, the distinct internal and external lamina, and the pattern of neuron interconnection. Analogous to the bits in an SR flip-flop, two arbitrary images can hold each other in place in an association processor and thereby form a short-term image memory. Overlay masks can focus attention on specific image regions. Spherically symmetric wavelets, identical to those found in the receptive fields of the retina, enable efficient image computations. Stability and noise reduction in reciprocal continuous wavelet transform representations can be achieved using an orthogonal projection based on the reproducing kernel.