Electrical Engineering and Systems Science > Image and Video Processing
[Submitted on 31 Jan 2019 (v1), revised 4 Feb 2019 (this version, v2), latest version 17 Feb 2019 (v4)]
Title:Image Reconstruction from Undersampled Confocal Microscopy Data using Multiresolution Based Maximum Entropy Regularization
View PDFAbstract:We consider the problem of reconstructing 2D images from randomly undersampled confocal microscopy samples. The well known total variation regularization, which is the L1 norm on the magnitude of derivatives, turns out to be unsuitable for this problem; it is unable to handle both noise and undersampling together because the measurement model is spatially localized. This issue is linked with notion of phase transition phenomenon observed in compressive sensing research, which essentially is the break-down of total variation methods, when sampling density gets lower than certain threshold. The severity of this breakdown is determined by the so-called mutual incoherence between the derivative operators and measurement operator. In our problem, the mutual incoherence is low, and hence the total variation regularization gives serious artifacts in the presence of noise even when the sampling density is not very low. In our recent publication (JOSA A, vol. 35, issue 10), we proposed a multi-resolution based regularization that performs better than total variation regularization. Here, we propose a further significantly improved multi-resolution based regularization approach that is better than the previous form in following ways: (i) the proposed form has a better adaptability to the image structure since it is based on image Hessian containing the orientation of image features; (ii) it is constructed such that the corresponding prior probability distribution has highest entropy meaning that the distortion caused by the regularization is minimum. We demonstrate the superiority of the proposed regularization using several reconstruction examples.
Submission history
From: Muthuvel Arigovindan [view email][v1] Thu, 31 Jan 2019 20:36:27 UTC (6,271 KB)
[v2] Mon, 4 Feb 2019 10:38:23 UTC (6,234 KB)
[v3] Wed, 6 Feb 2019 04:15:09 UTC (1,575 KB)
[v4] Sun, 17 Feb 2019 12:51:44 UTC (1,578 KB)
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.