Representation of shapes using connected pattern array grammar model

Abstract

From any image, the objects are recognized using patterns (or connected patterns). The ability to efficiently describe image patterns is of fundamental importance. Array Grammar models can be employed in generating patterns on a two — dimensional (2D) plane. Study of patterns generated by such grammar models form an important step in characterization and classification of images. Also, it is difficult to represent all connected patterns by proposing a model or pictorial representation. Basically a connected pattern of any neighborhood represents a shape. In this paper, we proposed a Connected Pattern Array Grammar (CPAG) model, a new model of array grammar capable of generating connected patterns. Generation of connected pattern on a 3 × 3 image neighborhood on a 2D plane is discussed as an example to demonstrate the capability of the proposed CPAG grammar model.