Content Based Image Retrieval Using Hybrid Technique

Content base image retrieval has become an active research area today. We present a Hybrid approach for content base image retrieval using the K-mean and Hierarchical clustering methods for better and efficient access of images. We design and implement hybrid clustering methods in JAVA. The proposed system demonstrated a promising and better retrieval method on a database containing random images according to some categories like clouds, landscape, fish and trees etc. Images are in JPEG, PNG and GIF format. Images are storing in backend database and fetching according to the attributes of images. To extract the features of content base image retrieval using Mean for location measures and Euclidean distance (ED) for distance measures. The performance of system has been evaluated using precision and recall methods.