Invention Grant
- Patent Title: Correlatability analysis for sparse alignment
- Patent Title (中): 稀疏对齐的相关性分析
-
Application No.: US12332251Application Date: 2008-12-10
-
Publication No.: US08351740B2Publication Date: 2013-01-08
- Inventor: Tuan Quang Pham , Stephen James Hardy
- Applicant: Tuan Quang Pham , Stephen James Hardy
- Applicant Address: JP Tokyo
- Assignee: Canon Kabushiki Kaisha
- Current Assignee: Canon Kabushiki Kaisha
- Current Assignee Address: JP Tokyo
- Agency: Fitzpatrick, Cella, Harper & Scinto
- Priority: AU2007240236 20071211
- Main IPC: G06K9/00
- IPC: G06K9/00

Abstract:
Disclosed herein is a method of estimating a geometrical relationship between a first image (101) and a second image (102), wherein the second image (102) includes a noise component. The method determines a location and size of each one of a plurality of image patches (201), based on the noise component included in the second image (102) and correlation information derived from the first image (101). The method then identifies a plurality of first image areas in the first image and a corresponding plurality of second image areas in the second image, based on the location and size of each one of the plurality of image patches. Each first image area of the first image (101) corresponds to a related second image area of the second image (102). The method then determines a geometrical relationship between the first and second images (101, 102) by comparing, for each one of the first image areas, information located within the first image area with information located within the corresponding related second image area.
Public/Granted literature
- US20090148051A1 CORRELATABILITY ANALYSIS FOR SPARSE ALIGNMENT Public/Granted day:2009-06-11
Information query