Normal and reverse order serial models for huge image database based on Formal Concept Analysis

In order to mine information effectively in the huge image database, we propose normal and reverse order serial models for two attribute partial order structure diagrams generated by the different attribute group for the same object group. And through a simple example using huge video/image database randomly selected from internet, it is confirmed the effectiveness and brevity of the proposed method, to provide a new method of image mining and to discover more hierarchical relationship.