Image Object Retrieval and Recognition Based On Descriptive Features Using SVM

Content Based Image Retrieval is a most emerging technique which helps us to retrieve the image from the large scale of image database base on the query image. In this paper, an image object retrieval system using the low-level features of image sub regions is proposed. The descriptive features namely color, shape and texture of the images are computed from the histograms of the quantized color space using color coherence vector, Gabor Filter and Gray Level Co-occurrence Matrix respectively. The N-class SVM is used to classify and labeling the images. Similarity metric between the query and target image is calculated using Euclidean distance measure. Experimental results show that the proposed method provides better retrieving result than existing methods.