Patent · US Expired

Wavelet-based clustering method for managing spatial data in very large databases

US6882997B1 · kind B1 · utility

35Cited by
8References
3Claims
0Family size

Assignee

Inventors

Key dates

Filing dateAug 24, 2000
Grant dateApr 19, 2005
Priority date
Expiry dateApr 14, 2021

Classification

  • Technology area (CPC Y)Emerging Cross-Sectional Technologies
  • CPC primaryY10S707/99945
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

The method termed WaveCluster for mining spatial data. WaveCluster considers spatial data as a multidimensional signals and applies wavelet transforms, a signal-processing technique, to convert the spatial data into the frequency domain. The wavelet transforms produce a transformed space where natural clusters in the data become more distinguishable. The method quantizes a feature space to determine cells of the feature space, assigns objects to the cells, applies a wavelet transform on the quantized feature space to obtain a transformed feature space, finds connected clusters in sub bands at different levels of the transformed feature space, assigns labels to the cells, creates a look-up table, and maps the objects to the clusters. The method can manage spatial data in a two-dimensional feature space. The method also is applicable to a feature space that is made up of an image taken by a satellite.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.