Improved Lower Content Feature based Content Based Image Retrieval using Support Vector Machine

The growth of internet technology increase the rate of data and expectation of searching and retrieval of data are very high. In such scenario Content Based Image Retrieval play very important role. The power of content based image retrieval is features description of image for better retrieval of image. In this paper we discuss a novel technique for content based image retrieval based on lower content of features such as color and texture. For the extraction of lower content features of image use MPEG-7 features descriptor. For the Extracted feature classified by supervised support vector machine classifier. Here we cascaded our classifier with RBF kernel function and improved the rate of classification. And this improved classification improved the features retrieval of content based image retrieval. In feature selection, lower features, mel-cepstral features and their combinations are considered for the task. While lower features like color, texture and sub band energies capture the spectral characteristics of the image, some of characteristic features of image are lost. Our cascaded support vector machine classifier reduced the negative query and improved the performance of content based image retrieval system.