Open Access
Translator Disclaimer
2005 Finding Minimal and Maximal Sets of Spatial Relationships in Pictorial Retrieval Systems
Qing-Long Zhang, Stephen S.-T. Yau
Commun. Inf. Syst. 5(3): 311-340 (2005).


Spatial reasoning is an important component in pictorial retrieval systems. There are two approaches to handling spatial relationships: the well-known one is to use algorithms on which most earlier work such as [13, 17, 21] is based, and the recent one [30] is to construct deductive rules that allow spatial relationships to be deduced. Sistla et al. [30] developed a system of rules $\cal R$ on reasoning about basic spatial relationships that are of common interest in pictorial databases. In this paper, we consider the following two problems with that system of rules $\cal R$: the deduction problem (that is, to deduce new spatial relationships from a given set $F$ of spatial relationships) and the reduction problem (that is, to eliminate redundant spatial relationships from $F$. We use the mathematically simple matrix representation approach to show that these two problems can be solved by efficient (i.e., polynomial-time) algorithms. The time required by both of them is at most a constant multiple of the time to compute the transitive reduction of a directed graph with $n$ vertices or to compute the transitive closure of a directed graph with $n$ vertices or to perform $n\times n$ Boolean matrix multiplication, and thus is always bounded by time complexity $O(n^3)$ (and space complexity $O(n^2)$), where $n$ is the number of all involved objects.


Download Citation

Qing-Long Zhang. Stephen S.-T. Yau. "Finding Minimal and Maximal Sets of Spatial Relationships in Pictorial Retrieval Systems." Commun. Inf. Syst. 5 (3) 311 - 340, 2005.


Published: 2005
First available in Project Euclid: 8 June 2006

zbMATH: 1110.68380
MathSciNet: MR2230530

Rights: Copyright © 2005 International Press of Boston


Vol.5 • No. 3 • 2005
Back to Top