当前位置: 首页 / 科研学术 / 学术预告 / 正文

PDEs and Image decomposition -based methods for various types of optical interferometry fringe images with poor quality

作者:   时间:2018-10-17   点击数:

报告题目:PDEs and Image decomposition -based methods for various types of optical interferometry fringe images with poor quality

报告人:唐晨教授,天津大学

时间:2018年10月18日 10:30-11:30

地点:知新楼B1044

摘 要:Optical interferometry fringes have various types.According to the density values, optical fringe images can be divided into: low-density fringe images, high-density fringe images, variable density fringe images. According to the continuity, optical fringe images can be divided into continuous fringe images and discontinuous fringe images. This report will introduce PDEs and Image decomposition -based methods for various types of optical interferometry fringe images with poor quality

Mainly includeing:

(1) Adaptive oriented PDEs based on new controlling speed function for discontinuous optical fringe patterns;

(2) A shape-preserving oriented PDEs based on a new fidelity term;

(3) Two parabolic-hyperbolic oriented PDEs for Improving computational efficiency;

(4) Fuzzy c-means clustering based discontinuities identification and the filtering method for discontinuous ESPI fringe patterns;

(5) General filtering method for ESPI fringe images with various densities based on variational image decomposition(VID)

(6) image decomposition model Shearlet-Hilbert-L2

(7) VID_based for estimation of orientation and density of optical interferometry fringe with variable density;

(8) VID_based for skeletonization of optical interferometry fringe with variable density.

In addition. This report will introduce our recent works on learning machine.

唐晨教授简介:Ms. Tang Chen is a Professor, doctoral tutor, works in School of electronic information engineering, Tianjin University. Her first research interest is the modern optical testing technology and application; The second research interest is the extraction of optical information based on optimization theory and application. In recent years, Prof. Tang hosted three National Natural Science Foundations, and a number of provincial and ministerial scientific research projects. Also her group has presented many advanced methods for the extraction of optical information. For example, her group has presented various types of anisotropic partial differential equation filtering methods for low quality electronic speckle interference fringe patterns with different density based on optimization theory. The proposed methods are able to perform impressively in speckle noise suppression and fringe preservation. Also her group proposed the gradient vector field methods for skeleton extraction of electronic speckle interference fringe patterns. The most main advantage of this method is its simplicity. It works directly on gray-scale images without the segmentation stage. In recent years she has published 43 papers, among them, the number of SCI included papers is 36. These papers have been cited by SCI source journals for nearly 260 times. And these papers have been published in Optics Letters, Optics Express, Applied Optics. IEEE Transactions on image processing and so on. One of her papers (published in Optics Letters, 2008) has been cited 61 times.

邀请人: 付树军教授

地址:中国山东省济南市山大南路27号   邮编:250100  

电话:0531-88364652  院长信箱:sxyuanzhang@sdu.edu.cn

Copyright@山东大学数学学院

微信公众号