Handling uncertain data in subspace detection

作者：

Highlights：

• A general solution for detecting data alignments (subspaces) in unordered multidimensional data.

• Our approach can handle both exact data as well as data containing Gaussian distributed uncertainties.

• Our approach allows the concurrent detection of different types of data alignments.

• Our approach allows the detection of data alignments in heterogeneous datasets.

摘要

Highlights•A general solution for detecting data alignments (subspaces) in unordered multidimensional data.•Our approach can handle both exact data as well as data containing Gaussian distributed uncertainties.•Our approach allows the concurrent detection of different types of data alignments.•Our approach allows the detection of data alignments in heterogeneous datasets.

论文关键词：Hough transform,Uncertain data,Subspace detection,Shape detection,Grassmannian,Geometric algebra,Parameter space

论文评审过程：Received 21 June 2013, Revised 4 April 2014, Accepted 9 April 2014, Available online 19 April 2014.

论文官网地址：https://doi.org/10.1016/j.patcog.2014.04.013