Recognition of two-dimensional objects using geometric constraints based on a modified genetic algorithm

Abstract

The aim of the research work is to develop a recognition system for identifying and localising of a two- dimensional object from a scene image acquired using a standard CCD camera. Before proposing an effective matching paradigm, a survey on 2D object recognition systems is conducted to identify their strengths and weaknesses. To accomplish the 2D interpretation problem, a novel matching paradigm based on a modified genetic algorithm (GA) [35] [36] [37] is proposed and implemented for recognising 2D objects subjected to Euclidean transformation, partial occlusion and clusters of irrelevant objects.