Naslov
Hybrid Optimization Method and Algorithms for Monochrome Images Tone Approximation with Implementation

Autori
Aghajanyan, Albert ; Neydorf, Rudolf ; Vučinić, Dean

Vrsta, podvrsta i kategorija rada
Poglavlja u knjigama, znanstveni

Knjiga
Advances in Visualization and Optimization Techniques for Multidisciplinary Research

Urednik/ci
Vucinic, Dean ; Rodrigues Leta, Fabiana ; Janardhanan, Sheeja

Izdavač
Springer

Grad
Singapur

Godina
2019

Raspon stranica
317-356

ISBN
978-981-13-9805-6

ISSN
2195-4356

Ključne riječi
Hybrid Optimization ; Monochrome Images ; Tone Approximation

Sažetak
The chapter considers the Monochrome Multi-tone Images (MMI) Tone Approximation (TA) problem. The TA procedure consists in reducing the image single-color tones palette size by replacing the original tones values with the approximated ones. The main problem is the selection of the appropriate approximation tones ; in other words, there is a need to define the optimal palette. To provide optimal TA for monochrome images a hybrid algorithm is developed and implies a 2- stage MMI processing. In the first stage the modified evolutionary-genetic algorithm is used. The main goal of the first stage is reducing the search area for the optimal approximation palette. In the second stage, the simple, but effective deterministic algorithm scans the nearest neighbourhood of the suboptimal solution, which was found in the first stage. The scanning of the nearest neighbourhood guarantees that the found extreme approximation palette is fulfilling the optimization criterion and that it is sub-optimized in respect to the total TA processing time.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo, Interdisciplinarne tehničke znanosti

POVEZANOST RADA