Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement Learning

Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement Learning

Bibliographic Details
Main Author: Vassilo, Kyle
Language:English
Published: University of Dayton / OhioLINK 2020
Subjects:
Artificial Intelligence
Computer Engineering
Electrical Engineering
deep learning
super resolution
generative adversarial networks
infrared imaging
reinforcement learning
asynchronous advantage actor-critic
a3c
Online Access:http://rave.ohiolink.edu/etdc/view?acc_num=dayton1606146042238906
id ndltd-OhioLink-oai-etd.ohiolink.edu-dayton1606146042238906
record_format oai_dc
spelling ndltd-OhioLink-oai-etd.ohiolink.edu-dayton16061460422389062021-08-03T07:16:31Z Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement Learning Vassilo, Kyle Artificial Intelligence Computer Engineering Electrical Engineering deep learning super resolution generative adversarial networks infrared imaging reinforcement learning asynchronous advantage actor-critic a3c Recent studies have shown that Deep Learning (DL) algorithms can significantly improve Super Resolution (SR) performance. Single image SR is useful in producing High Resolution (HR) images from their Low Resolution (LR) counterparts. The motivation for SR is the potential to assist algorithms such as object detection, localization, and classification. Insufficient work has been conducted using Generative Adversarial Networks (GANs) for SR on infrared (IR) images despite its promising ability to increase object detection accuracy by extracting more precise features from a given image. This work adopts the idea of a relativistic GAN that utilizes Residual in Residual Dense blocks (RRDBs) for feature extraction, a novel residual image addition, and a Pixel Transposed Convolutional Layer (PixelTCL) for up-sampling. Recent work has validated the use of GANs for Visible Light (VL) images, making them a strong candidate. The inclusion of these components produce more realistic and natural features while also receiving superior metric values. Supplemental research applies a multi-agent Reinforcement Learning (RL) algorithm to Single Image Super-Resolution (SISR), creating an advanced ensemble approach for combining powerful GANs. In our implementation each agent chooses a particular action from a fixed action set comprised of results from existing GAN SISR algorithms to update its pixel values. The pixel-wise arrangement of agents and rewards encourages the algorithm to learn a strategy to increase the resolution of an image by choosing the best pixel values from each option. 2020 English text University of Dayton / OhioLINK http://rave.ohiolink.edu/etdc/view?acc_num=dayton1606146042238906 http://rave.ohiolink.edu/etdc/view?acc_num=dayton1606146042238906 unrestricted This thesis or dissertation is protected by copyright: all rights reserved. It may not be copied or redistributed beyond the terms of applicable copyright laws.
collection NDLTD
language English
sources NDLTD
topic Artificial Intelligence
Computer Engineering
Electrical Engineering
deep learning
super resolution
generative adversarial networks
infrared imaging
reinforcement learning
asynchronous advantage actor-critic
a3c
spellingShingle Artificial Intelligence
Computer Engineering
Electrical Engineering
deep learning
super resolution
generative adversarial networks
infrared imaging
reinforcement learning
asynchronous advantage actor-critic
a3c
Vassilo, Kyle
Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement Learning
author Vassilo, Kyle
author_facet Vassilo, Kyle
author_sort Vassilo, Kyle
title Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement Learning
title_short Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement Learning
title_full Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement Learning
title_fullStr Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement Learning
title_full_unstemmed Single Image Super Resolution with Infrared Imagery and Multi-Step Reinforcement Learning
title_sort single image super resolution with infrared imagery and multi-step reinforcement learning
publisher University of Dayton / OhioLINK
publishDate 2020
url http://rave.ohiolink.edu/etdc/view?acc_num=dayton1606146042238906
work_keys_str_mv AT vassilokyle singleimagesuperresolutionwithinfraredimageryandmultistepreinforcementlearning
_version_ 1719457957496750080

Similar Items