Neuro-Fuzzy based Image Retrieval System with Improved Shape and Texture Features

A generalized Neuro-Fuzzy based Content Based Image Retrieval (CBIR) system is proposed. The system is trained for colour, texture and shape features using General Fuzzy Min-Max Neural Network (GFMNN). Flexibility and robustness is achieved by accepting any number and types of different input features as well with the concept of class labels assigned for each hyperbox. The existing architecture is simplified and the system is trained in pure clustering mode which helps in reducing the computational complexity. By controlling user parameters the system can categorize images as per the users need. With modified texture and shape features combined with colour features, the proposed CBIR system gives an efficient automated retrieval of similar images.