Real Time Object Recognition System

摘要

This paper introduces a real time object recognition system. This novel algorithm learns to classify objects into their respective classes. The proposed approach is a combination of traditional image processing techniques, and modern artificial intelligence techniques. Artificial intelligence approach of fuzzy logic and artificial neural networks is used for image enhancement and classification purpose respectively. Image segmentation and feature extraction are computed using traditional techniques of seeded region growing and Zernike moments respectively. The proposed approach is implemented in MATLAB, and real time images are captures for training and testing the proposed system. The proposed algorithm produced accurate results and perfectly recognized all the objects, irrespective of their size, position and orientation. This approach can be easily used in applications which require real time object recognition.