Feature Extraction and Detection of Simple Objects Using Particle Swarm Optimisation

The purpose of this paper is to demonstrate the application of particle swarm optimisation to the detection of simple objects. The paper’s new contribution to object detection is application of particle swarm optimisation for extraction of geometric properties of an object in an image for accurate recognition especially in noisy environments. In this approach, the edges and the corners of an object are detected by particle swarm optimisation algorithm and then the object is classified based on number of the corners and attributes of the edges by a simple fuzzy rule-based classifier. Several simple geometric objects in different aspects have been used in the variety of noise levels for testing of the system. This system can categorise images containing these simple objects even with high noise levels more accurately in contrast to other approaches proposed in the last literature.