Content based image retrieval (CBIR) system allows users to retrieve similar images from a large image database based upon a query image. Furthermore, the challenges faced by content-based image retrieval system are the semantic gap and perception subjectivity. In this paper, a combined feature fuzzy approach based CBIR system for the retrieval of bio-medical images is proposed in order to provide better image classification and fast image retrieval. In the proposed method the texture features like contrast, mean, standard deviation, energy and entropy are extracted from the image and converted into fuzzy aggregation membership functions. The result shows that the proposed approach retrieves images efficiently and accurately than the other existing techniques.