With the development of the World-Wide Web, there has been an explosion in the volume of multimedia database. In order to search desired information on such huge and quick changing internet, more exact and faster retrieval tools should be developed. In this paper, a new image retrieval technique based on concept lattices is presented, named Concept Lattices- Based Image Retrieval, lattice browsing allows one to reach a group of images via one path. Because it can produce a lot of redundancies attributes when constructing the concept lattices by using a general method, this paper also proposes a method of attribute reduction of concept lattices based on discernibility matrix and boolean calculation to reduce the context of concept lattices. The scale of the problem is reduced by using this method. At the same time, the efficiency of image retrieval is improved, which is reflected in our experiment.