Recognition of Overlapped 2D Geometrical Objects

Abstract

In this paper the main focus is to study the techniques to detect or recognize the geometrical objects which are overlapped or merged with each other and to find out the best possible way to determine their respective shapes. Object detection is the process or technique of finding occurrence of real-world objects such as faces, objects, bicycles, and buildings in images or videos. Algorithms used for object detection typically use extracted features and learning algorithms to detect instances of an object division. It is commonly used in different applications such as image retrieval, security and automated vehicle parking systems. In meantime there are various different techniques available to detect object of a particular geometric shape from 2D images. But they are not much reliable techniques that identify features of objects of an image and recognize the object having geometric shape like circle, square, rectangle and triangle and other shapes. Object detection plays an important role in image processing, It helps in to identify any particular object .Object detection is basically used to identify an individual object from number of objects in an image. The proposed system includes a new algorithm to separate the touching and overlapping circles edges based on the radius range. Initial attempt include finding circles by increasing detection sensitivity and final step, include finding the dark and bright circle edges by lowering the values of edge threshold.