A Method Using Morphology and Histogram for Object-based Retrieval in Image and Video Databases
Thi Thi Zin, Hiromitsu Hama · 2007
We propose a method for object-based retrieval in image and video databases. In doing so, both color and shape information is taken into account to enrich the querying capabilities for object-based retrieval. The morphological operations are performed on each uniform color region. We use three kinds of histograms: color histogram, distance histogram, and angle histogram. It is shown that the proposed approach is invariant to scaling, rotation, and translation. Finally, performance experiments are carried out by the evaluation of the effectiveness and the robustness of the proposed method. Key words: Object-based retrieval, Uniform color region, Image and video