Proceedings of the 2016 International Conference on Computer Science and Electronic Technology

Adaptive Iteration Stopping Criterion for AMSS Equations

Authors
Changle Li, Gangfeng Liu, Jie Zhao
Corresponding Author
Changle Li
Available Online August 2016.
DOI
https://doi.org/10.2991/cset-16.2016.2How to use a DOI?
Keywords
Partial differential equation, AMSS equation, cross entropy, image filter, surface defects
Abstract
When AMSS (Affine Morphological Scale Space) operator is applied in image filtering, the scale parameter has great impact on the filtering results. In order to determine the parameters more precisely, this paper analyzed affine invariance properties and classical invariance properties of AMSS operator using the affine transform theory And the numerical solution of AMSS equation is realized with the finite difference method . Based on the cross entropy theory, the forms of cross entropy under the two situations of standard image and nonstandard image were analyzed. The relationship between cross entropy and the scale parameter was explored in both cases, and the AMSS equation's iteration stopping time is determined based on this criterion. Experiment shows that the satisfactory smooth images can be achieved based on the method of cross entropy iteration stopping, under the circumstances of both standard and nonstandard images.
Open Access
This is an open access article distributed under the CC BY-NC license.

Download article (PDF)

Proceedings
2016 International Conference on Computer Science and Electronic Technology
Part of series
Advances in Computer Science Research
Publication Date
August 2016
ISBN
978-94-6252-213-8
ISSN
2352-538X
DOI
https://doi.org/10.2991/cset-16.2016.2How to use a DOI?
Open Access
This is an open access article distributed under the CC BY-NC license.

Cite this article

TY  - CONF
AU  - Changle Li
AU  - Gangfeng Liu
AU  - Jie Zhao
PY  - 2016/08
DA  - 2016/08
TI  - Adaptive Iteration Stopping Criterion for AMSS Equations
BT  - 2016 International Conference on Computer Science and Electronic Technology
PB  - Atlantis Press
SN  - 2352-538X
UR  - https://doi.org/10.2991/cset-16.2016.2
DO  - https://doi.org/10.2991/cset-16.2016.2
ID  - Li2016/08
ER  -