In CBIR main problem is to extract the image features that effectively represent the contents of an image in a database which may be difficult to use, using single feature. Whereas with combination of different features it would give efficient results. The different combination used maybe color-edge; color-shape-texture; color-texture; shape-texture. With more features different aspects of an image could be represented. In all the methods the texture is characterized by the statiscal distribution of the image intensity while the shape features are described with the segmentation of the region or the objects. The combination of CT has faster retrieval method which is robust to scale and translate objects in an image. Also the single feature like Gabor features has less precision than combination features of color moments and the Gabor texture. Also the experiments have proved to increase the efficiency of results.