In this paper, a new spatial knowledge representation model named"Two Dimension Begin-End Boundary String" (2D Bε-string) is proposed. The 2D Bε-string represents an icon by its MBR boundaries and a nu mber of "dummy objects". The 2D Bε-string can intuitively and natura lly represent the pictorial spatial information without any spatial operator. In addition, an image similarity evaluation method based on the modified "Longest Common Subsequence" (LCS) algorithm is presented. By the proposed evaluation method, not only those images which all of the icons and their spatial relationships fully accord with the query image can be sifted out, but also those images which partial icons and/or spatial relationships are similar to the query image can be applied to. It resolves the problems that the query targets and/or spatial relationships are not certain. Our representation model and similarity evaluation also simplify the retrieval progress of linear transformations, including rotation and reflection of an image.
Proceedings of the 2000 ICS:Workshop on Image Processing and Pattern Recongnition，頁125-132