Multithreading Image Processing in Single-core and Multi-core CPU using Java

Abstract

Multithreading has been shown to be a powerful approach for boosting a system performance. One of the good examples of applications that benefits from multithreading is image processing. Image processing requires many resources and processing run time because the calculations are often done on a matrix of pixels. The programming language Java supports the multithreading programming as part of the language itself instead of treating threads through the operating system. In this paper we explore the performance of Java image processing applications designed with multithreading approach. In order to test how the multithreading influences on the performance of the program, we tested several image processing algorithms implemented with Java language using the sequential one thread and multithreading approach on single and multi-core CPU. The experiments were based not only on different platforms and algorithms that differ from each other from the level of complexity, but also on changing the sizes of the images and the number of threads when multithreading approach is applied. Performance is increased on single core and multiple core CPU in different ways in relation with image size, complexity of the algorithm and the platform.