Figure 2: Block diagram of the proposed technique based on visual words fusion of SIFT and LIOP descriptors.