Signal-to-Symbol Transformation and Vice Versa: From Fundamental Processes to Representation.

Reasoning in the physical world, characterized by artificial agents, perceived senses and signals, is performed on symbolic entities. The acute problem of reasoning in this process is the determination of the symbolic element, or the segmentation of the perceptive signal, termed as segmentation problem in the domain of computer vision. It has been assumed that separation of perception from action is not an acute possibility. However, more focused studies have revealed that perception-action cycle is governed by behavior, or a unit of complex behaviors. Most low-level behaviors simply represent a perception-action connection, without the need for symbolic representation. An agent that carries out the task of symbol-to-signal transformation must possess the capability of invoking the proper behavior based on this transformation. The emerging mathematical field of hybrid systems is one such tool. Researchers have recognized the significance of hybrid systems in domains like computer vision, and artificial intelligence, and it promises to enhance its impact in these domains.