Optimizacija metode kompresije slika bez gubitaka za platformu Android

master's thesis

• Title (english): Lossless Image Compression Method Optimization for Android Platform
• Author: Ivan Blažeković
• Mentor: Josip Knezović (mentor)
• Committee member: Josip Knezović (predsjednik povjerenstva)
• Committee member: Hrvoje Mlinarić (član povjerenstva)
• Committee member: Igor Čavrak (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-07-18, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu analizirano je vrijeme izvođenja pojedinih metoda za kompresiju slika bez gubitaka temeljenih na tehnikama predviđanja. U postojeću klijentsku aplikaciju integrirana je podrška za slikovne podatke formata DICOM. DICOM (Digital Imaging and COmmunications in Medicine) je najrašireniji i najprihvaćeniji standard za slike u medicini. Za integraciju je korištena vanjska biblioteka Imebra, koja sadrži sve potrebne funkcionalnosti za manipuliranje slikovnim podacima DICOM formata. Nakon početnog profiliranja idetificirano je nekoliko točaka koje su analizirane i na temelju analize postignuta je optimizacija izvođenja. U aplikaciji je integrirana podrška za Android NDK - skup alata koji omogućuje da se dio aplikacije implementira koristeći programske jezike sa direktnim prevođenjem poput C-a ili C++-a. Ispitana je i integrirana je podrška za ARM NEON intrinzike. NEON tehnologija je napredna SIMD(single instruction multiple data)arhitektura za procesore serije ARM Cortex-A. Provedeno je profiliranje optimirane i originalne aplikacije nad skupom ispitnih slika te je postignuto ukupno prosječno ubrzanje izvođenja od 3.38 puta.

Abstract (english)

In this thesis,performance time of lossless image compression techniques based on
predictive coding is analyzed. Support for DICOM image format has been integrated within the existing client application. DICOM (Digital Imaging and COmmunications in Medicine) is the most widely used and accepted standard for medical imaging. Imebra library has been used for integration, which contains all the necessary functionality for manipulating DICOM image data. After initial profiling, several points for optimization were identified and analyzed. Based on that analysis the performance was optimized. The application has integrated support for Android NDK.NDK is a set of tools that enables part of the application to be implemented using native language code scripts such as C or C ++. It was established by testing that using the NDK for the selected method did not achieve optimization of performance. Support for ARM NEON intrinsics has been tested and integrated. NEON technology is an advanced single-instruction multiple data architecture for the ARM Cortex-A processor series. Profiling of the optimized and original application over the set of test images was done, and in overall, an average acceleration of 3.38 times was achieved.

• Keywords: kompresija slike bez gubitaka; metode predviđanja; optimizacija; Android OS; Android NDK; ARM NEON
• Keywords (english): lossless image compression; predictive coding; optimization; Android OS; Android NDK; ARM NEON
• Language: croatian
• URN:NBN: urn:nbn:hr:168:897284
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 19:08:17