Quite OK (Entropy Coded) Image Compression

Lossless image compression is a subfield of computer science that aims to reduce the size of digital images while preserving all the original information. In a raster image, this is typically achieved by retaining the exact bitmap values of individual pixels. Compression allows images to be more easily stored, transmitted, and shared. A large number of digital images today are written in the Portable Network Graphics (PNG) format due to its free and widespread support. In 2021, developer Dominic Szablewski at Phoboslab developed and announced a new lossless image compression algorithm called the “Quite OK Image Format” (QOI), which promised to achieve similar compression ratio to that of PNG while taking significantly less time to encode and decode.This thesis contributes in three ways. Firstly, we provide a detailed description of the QOI format and its official implementation in detail, analyzing its effectiveness and weaknesses. Secondly, we propose a series of changes that trade encoding speed for a better compression ratio. Lastly, we benchmark the official reference implementation of QOI and our implementation against the reference implementation of PNG, evaluating their performance differences.

Author

• Qiu, Chuhui

Language

• en

Date created

• 2023

Type of Work

• Master's Thesis

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Computer Science

Degree

• M.S.

Advisor

• Pless, Robert

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/pz50gw91c