This paper provides the analysis of content based image retrieval (CBIR) system. The previous results indicate that there are lot of works in the direction of CBIR are already done but the work is going on for the hunt of improving the retrieval precision. In this paper dominant, local binary pattern (LBP) and color moment (CM) have been hybridized with the combination of colour, shape and texture features for efficient image retrieval system. First the dominant colors have been extracted and then LBP patterns are vectored. The combination is then process to CM and then index vector are calculated based on the color, shape and texture features. The results obtained clearly indicate that it is better than LBP, CM and LBP and CM in combination.