A Stepwise Similarity Approximation of Spatial Constraints for Image Retrieval

QL Zhang,SST Yau
DOI: https://doi.org/10.1117/12.632728
2005-01-01
Abstract:A real image is assumed to be associated with some content-based meta-data about that image (i.e., information about objects in the image and spatial relationships among them). Recently Zhang and Yau have addressed the approximate picture matching problem, and have presented a stepwise approximation of intractable spatial constraints in an image query. In particular, in contrast with very few cases done in earlier related works, Zhang-Yau's algorithmic analysis shows that there are all possible 16 cases for results of the object matching step of image retrieval, and 13 out of these 16 cases are valid for the stepwise approximation of spatial constraints while the only other 3 cases axe identified impossible for finding an exact picture-matching between a query picture and a database picture. In this paper, Zhang and Yau have successfully used the stepwise approximation method to work out a similarity measure between a query image and a database image, for image retrieval. The proposed similarity measure utilizes the similarity measures previously developed, by Gudivada and Raghavan (1995) and El-Kwae and Kabuka (1999), for the scenario of the single occurrence of each object in both query and database images, and extends to cover all 13 valid cases.
What problem does this paper attempt to address?