Fuzzy C-Means Initialized by Fixed Threshold Clustering for Improving Image Retrieval

Fuzzy C-Mean (FCM) algorithm is one of the well-known unsupervised clustering techniques. Such an algorithm can be used for unsupervised image clustering. Then, images can be indexed in databases. The different initializations cause different evolutions of the algorithm. Random initializations may lead to improper convergence. This paper proposes FCM initialized by fixed threshold clustering. The case study regards to retrieve from the database the color JPEG images, indexed by color histogram vectors. The result shows that the proposed method gives more accurate results than FCM with random initialization and color histogram clustering do.