Digital image processing through parallel computing in single-core and multi-core systems using MATLAB

Autor: Ritesh Reddy, Neeharika Kompala, Nitin Ravi
Rok vydání: 2017
Předmět:
Zdroj: 2017 2nd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT).
DOI: 10.1109/rteict.2017.8256639
Popis: Matlab by default is a single threaded application. so the single thread uses just one core irrespective of how many cores are available to run any simulation. Data analysis and visualization tool which has been designed with powerful support for matrices and matrix operations. when we talk here about digital image processing we basically refer to how we change the nature of the image either to improve its pictorial representation for better human understanding or to more suitable for machine interpretation. huge matrix calculations take place in matrix pixels, all these calculations are done in single core. here, we talk about three major digital image processing features provided by Matlab and how parallel computing feature provided by matlab helps its computational speed by decreasing the time of it execution. we compare matlab's performance on different systems using different number of cores. Overall the parallel computing tool in matlab helps in solving computationally intensive and data-intensive problems with much more ease and efficiency with regards to time.
Databáze: OpenAIRE