First Advisor

Feng Liu

Term of Graduation

Summer 2022

Date of Publication


Document Type


Degree Name

Master of Science (M.S.) in Computer Science


Computer Science





Physical Description

1 online resource (xi, 65 pages)


Neural networks are a powerful machine learning tool, especially when trained on a large dataset of relevant high-quality data. Generative adversarial networks, image super resolution and most other image manipulation neural networks require a dataset of images and matching target images for training. Collecting and compiling that data can be time consuming and expensive. This work explores an approach for building a dataset of paired document images with a matching scanned version of each document without physical printers or scanners. A dataset of these document image pairs could be used to train a generative adversarial network or image super resolution neural network to convert a scanned document into a pristine document free of artifacts. It could also be used in optical character recognition of scanned documents to improve understanding of documents with degraded quality. Generating a dataset like this without mechanical hardware saves time and materials and has the potential to build similar paired image datasets for other applications. The proposed approach centers on conditional generative adversarial networks to generate the paired dataset from unpaired document images. This work explores StyleGAN2, CycleGAN, CUT, Pix2PixHD, SPADE and SEAN. I find that the base version of each model is currently insufficient for this task.


©2022 David Jonathan Hawbaker

In Copyright. URI: This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

Persistent Identifier