Patent · US Active

Attribute factor analysis method, device, and program

US10268876B2 · kind B2 · utility

1Cited by
1References
14Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJul 17, 2014
Grant dateApr 23, 2019
Priority date
Expiry dateNov 5, 2034

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06V40/172
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

This invention relates to an attribute factor analysis method, which is a method of analyzing a factor of an attribute based on a training sample set including training samples each being a combination of reference image data and attribute data associated with the reference image data. The attribute factor analysis method includes: a division step of dividing an image region of the reference image data constituting each training sample of the training sample set into parts of a predetermined sample size in a mesh shape; a model construction step of constructing a regression model by performing sparse regression analysis for each of the parts based on the reference sample set; a dependency calculation step of calculating, with use of the regression model, for each training sample of the training sample set, a dependency between an explanatory variable representing a feature quantity of reference image data on each part and an objective variable representing the attribute data, to obtain an attribute factor analyzed result; and a visualization step of visualizing the attribute factor analyzed result to produce the visualized attribute factor analyzed result.

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