Shape Recognition based on Features matching using Morphological Operations

This paper presents the implementation of method of shape recognition among different regular geometrical shapes using morphological operations. Many algorithms have been proposed for this problem but the major issue that has been enlightened in this paper is over segmentation dodging among different objects. After an introduction to shape recognition concept, we describe the process of extracting the boundaries of objects in order to avoid over segmentation. Then, a shape recognition approach is proposed. It is based on some mathematical formulae. Our new algorithm detects the shapes in the following cases when (i) There are distinct objects in the given image. (ii) The objects are touching in the given image. (iii) The objects are overlapping in the given image. (iv)One object is contained in the other in the given image. Then with the help of boundaries concentrate and shape properties, classification of the shapes is done.