| On convergence of the Horn and Schunck optical-flow estimation method. | |
| | |
MedLine Citation:
|
PMID: 15648874 Owner: NLM Status: MEDLINE |
Abstract/OtherAbstract:
|
The purpose of this study is to prove convergence results for the Horn and Schunck optical-flow estimation method. Horn and Schunck stated optical-flow estimation as the minimization of a functional. When discretized, the corresponding Euler-Lagrange equations form a linear system of equations We write explicitly this system and order the equations in such a way that its matrix is symmetric positive definite. This property implies the convergence Gauss-Seidel iterative resolution method, but does not afford a conclusion on the convergence of the Jacobi method. However, we prove directly that this method also converges. We also show that the matrix of the linear system is block tridiagonal. The blockwise iterations corresponding to this block tridiagonal structure converge for both the Jacobi and the Gauss-Seidel methods, and the Gauss-Seidel method is faster than the (sequential) Jacobi method. |
| | |
Authors:
|
Amar Mitiche; Abdol-Reza Mansouri |
Related Documents
:
|
22979474 - Using optimization for acoustic cloak design. 16593244 - Gravitational field of a charged mass point. 23655024 - Remembering one of maa dah-you's last essential contributions to acoustics. 16568774 - Does the rayleigh equation apply to evaluate field isotope data in contaminant hydrogeo... 10944334 - Towards automated laue data processing: application to the choice of optimal x-ray spec... 19792514 - Hexatic and mesoscopic phases in a 2d quantum coulomb system. |
Publication Detail:
|
Type: Comparative Study; Journal Article; Research Support, Non-U.S. Gov't |
Journal Detail:
|
Title: IEEE transactions on image processing : a publication of the IEEE Signal Processing Society Volume: 13 ISSN: 1057-7149 ISO Abbreviation: IEEE Trans Image Process Publication Date: 2004 Jun |
Date Detail:
|
Created Date: 2005-01-14 Completed Date: 2005-02-10 Revised Date: 2006-11-15 |
Medline Journal Info:
|
Nlm Unique ID: 9886191 Medline TA: IEEE Trans Image Process Country: United States |
Other Details:
|
Languages: eng Pagination: 848-52 Citation Subset: IM |
Affiliation:
|
Institut National de la Recherche Scientifique, INRS-EMT, Montreal QC H5A 1K6, Canada. mitiche@inrs-emt.uquebec.ca |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
Algorithms* Artificial Intelligence Cluster Analysis Image Enhancement / methods* Image Interpretation, Computer-Assisted / methods* Information Storage and Retrieval / methods Movement* Numerical Analysis, Computer-Assisted Pattern Recognition, Automated / methods* Reproducibility of Results Sensitivity and Specificity Signal Processing, Computer-Assisted Subtraction Technique* Video Recording / methods* |
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: Nonlinear prediction for Gaussian mixture image models.
Next Document: Constraining active contour evolution via lie groups of transformation.