A neural architecture for the symmetric-axis transform

The symmetric-axis transform is a process that dynamically encodes the space of a visual shape through self-interaction of its contours. It is generally simulated using computer algorithms. A neural architecture of this transformation is presented that is conceptually simple enough for a hardware implementation. Its architecture consists of a wave-propagating map, orientation-selective columns detecting wave pieces of specific orientation, and a coincidence map detecting the clash of two wave fronts. We illustrate its operation on partial contours extracted from gray-scale images.