Saved in:
Bibliographic Details
Main Authors: Varela, Luis G., Boucheron, Laura E., Sandoval, Steven, Voelz, David, Siddik, Abu Bucker
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.07796
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929578213113856
author Varela, Luis G.
Boucheron, Laura E.
Sandoval, Steven
Voelz, David
Siddik, Abu Bucker
author_facet Varela, Luis G.
Boucheron, Laura E.
Sandoval, Steven
Voelz, David
Siddik, Abu Bucker
contents The non-uniform blur of atmospheric turbulence can be modeled as a superposition of linear motion blur kernels at a patch level. We propose a regression convolutional neural network (CNN) to predict angle and length of a linear motion blur kernel for varying sized patches. We analyze the robustness of the network for different patch sizes and the performance of the network in regions where the characteristics of the blur are transitioning. Alternating patch sizes per epoch in training, we find coefficient of determination scores across a range of patch sizes of $R^2>0.78$ for length and $R^2>0.94$ for angle prediction. We find that blur predictions in regions overlapping two blur characteristics transition between the two characteristics as overlap changes. These results validate the use of such a network for prediction of non-uniform blur characteristics at a patch level.
format Preprint
id arxiv_https___arxiv_org_abs_2402_07796
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Estimation of non-uniform motion blur using a patch-based regression convolutional neural network (CNN)
Varela, Luis G.
Boucheron, Laura E.
Sandoval, Steven
Voelz, David
Siddik, Abu Bucker
Image and Video Processing
The non-uniform blur of atmospheric turbulence can be modeled as a superposition of linear motion blur kernels at a patch level. We propose a regression convolutional neural network (CNN) to predict angle and length of a linear motion blur kernel for varying sized patches. We analyze the robustness of the network for different patch sizes and the performance of the network in regions where the characteristics of the blur are transitioning. Alternating patch sizes per epoch in training, we find coefficient of determination scores across a range of patch sizes of $R^2>0.78$ for length and $R^2>0.94$ for angle prediction. We find that blur predictions in regions overlapping two blur characteristics transition between the two characteristics as overlap changes. These results validate the use of such a network for prediction of non-uniform blur characteristics at a patch level.
title Estimation of non-uniform motion blur using a patch-based regression convolutional neural network (CNN)
topic Image and Video Processing
url https://arxiv.org/abs/2402.07796