State-of-the-Art Approaches for Image Deconvolution Problems, including Modern Deep Learning Architectures

In modern digital microscopy, deconvolution methods are widely used to eliminate a number of image defects and increase resolution. In this review, we have divided these methods into classical, deep learning-based, and optimization-based methods. The review describes the major architectures of neura...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Micromachines (Basel) Jg. 12; H. 12; S. 1558
Hauptverfasser: Makarkin, Mikhail, Bratashov, Daniil
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Switzerland MDPI AG 14.12.2021
MDPI
Schlagworte:
Algorithms
Deconvolution
Deep learning
Design
Digital imaging
digital microscopy
Generative adversarial networks
Heterogeneity
image processing
Light
Microscopy
Neural networks
Noise
Optimization
Review
image processing
deconvolution
deep learning
digital microscopy
ISSN:2072-666X, 2072-666X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In modern digital microscopy, deconvolution methods are widely used to eliminate a number of image defects and increase resolution. In this review, we have divided these methods into classical, deep learning-based, and optimization-based methods. The review describes the major architectures of neural networks, such as convolutional and generative adversarial networks, autoencoders, various forms of recurrent networks, and the attention mechanism used for the deconvolution problem. Special attention is paid to deep learning as the most powerful and flexible modern approach. The review describes the major architectures of neural networks used for the deconvolution problem. We describe the difficulties in their application, such as the discrepancy between the standard loss functions and the visual content and the heterogeneity of the images. Next, we examine how to deal with this by introducing new loss functions, multiscale learning, and prior knowledge of visual content. In conclusion, a review of promising directions and further development of deconvolution methods in microscopy is given.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ObjectType-Review-3
content type line 23
ISSN:2072-666X
2072-666X
DOI:10.3390/mi12121558