Presentation | 2014-05-22 Window Size Estimation for Parametrically Describing Local Image Appearances by means of the Weighted Integral Method Hidetoshi GOTO, Hidekata HONTANI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The authors propose a new method that accurately estimates the parameters of the Gaussian function that describes a local appearance observed in each window at each location in given image. The Gaussian function is non-linear with respect to the parameters to be estimated, and the non-linearity makes the accurate parameter estimation difficult. In the proposed method, the weighted integral method is introduced to linearize the parameter estimation problem. The weighted integral method can estimate the parameters of the Gaussian functions that describe curvilinear structures. The estimated parameters, though, would be inaccurate if the local window includes more than one curvilinear structure. This means you need to adaptively select a window size that is appropriate for the estimation at each location in given images. In the proposed method, the window size is selected by evaluating the well-posedness of a system of the linear equations that the weighted integral method solves. Experimental results demonstrate that the proposed method estimates the parameters accurately with the appropriate size window. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | feature description / parameter estimation problem / scale analysis / weighted integral method |
Paper # | SIP2014-9,IE2014-9,PRMU2014-9,MI2014-9 |
Date of Issue |
Conference Information | |
Committee | MI |
---|---|
Conference Date | 2014/5/15(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | |
Vice Chair | |
Secretary | |
Assistant |
Paper Information | |
Registration To | Medical Imaging (MI) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Window Size Estimation for Parametrically Describing Local Image Appearances by means of the Weighted Integral Method |
Sub Title (in English) | |
Keyword(1) | feature description |
Keyword(2) | parameter estimation problem |
Keyword(3) | scale analysis |
Keyword(4) | weighted integral method |
1st Author's Name | Hidetoshi GOTO |
1st Author's Affiliation | Nagoya Institute of Technology() |
2nd Author's Name | Hidekata HONTANI |
2nd Author's Affiliation | Nagoya Institute of Technology |
Date | 2014-05-22 |
Paper # | SIP2014-9,IE2014-9,PRMU2014-9,MI2014-9 |
Volume (vol) | vol.114 |
Number (no) | 42 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |