ABSTRACT

This paper proposes a methodology for the content based image retrieval which is implemented on the cartoon images. The similarities between a query cartoon character image and the images in database are computed by the feature extraction using the fusion descriptors of SIFT (Scale Invariant Feature Transforms) and HOG (Histogram of Gradient). Based on the similarities, the cartoon images same or similar to query images are identified and retrieved. This method makes use of indexing technique for more efficient and scalable retrieval of the cartoon character. The experiment results demonstrate that the proposed method is efficient in retrieving the cartoon images from the large database.