Image Retrieval Based On Improved Growing Hierarchical Self-Organizing Map Clustering Aglorithm

With the development of the internet technology and multimedia technology a huge image information appeared. Text based image retrieval can’t satisfy people’s needs. Nowadays content-based image retrieval technology has become a new hotspot of image retrieval. The retrieval speed has been a bottleneck of image retrieval. Under the research of content-based image retrieve at home and abroad, a method that applies a hierarchical neural network clustering algorithm of machine learning to content-based image retrieve. This can improve the efficiency of image retrieve. The specific content is studied as follows:(1) Proposed an image feature extraction method which combines color histogram and imporved tamura texture extraction.(2) Presented a novel image retrieval method based on GHSOM clustering algorithm. The improved GHSOM algorithm applied AIC criterion to select proper growth parameter for each SOM map. Proper growth parameters make each map can represent data set well. AIC criterion was created according to the characteristics of the algorithm.(3) The traditional image retrieval needs to compute the distance between each picture in the database with the requested picture. The retrieval speed and accuracy is bad. Aiming at this drawback, this paper presented a novel image retrieval method based on improved GHSOM clustering algorithm. The first step of the image retrieval is to find out the clusters which one is most similar to the request picture in the GHSOM network. Then continue to retrieval the images in this cluster. The retrieval speed improved greatly. In the process of searching for similar cluster, the hierarchical structure of GHSOM network was made full use of to improve the retrieval efficiency.Experiments illustrate that the improved GHSOM algorithm can gain a better cluster performance. The speed of image matching has improved. Image retrieval accuracy has also been improved to some extent.