Image Representation Using Finite State Automata

Image representation using quadtrees is useful in many applications and is well known in computer graphics. It has been shown that we can represent quadtrees asrationallanguages that are recognizable by finite state automata (FSA) [1]. These FSA can use less memory and are more flexible than their corresponding quadtrees. Binary image representation using FSA follows these two main steps: (1) FSA construction based on quad trees and (2) FSA minimization [2]. This approach can be easily adapted to represent gray scale images, colors images or volume data (octrees). Image representation using FSA supports all quad tree operations (union, intersection, rotations, ... ) as well as all FSA-defined operations.