Parallel Generation of Image Layers Constructed by Edge Detection Using Message Passing Interface

Edge detection is one of the most fundamental algorithms in digital image processing. Many algorithms have been implemented to construct image layers extracted from the original image based on selecting threshold parameters. Changing theses parameters to get a high quality layer is time consuming. In this paper, we propose two parallel technique, NASHT1 and NASHT2, to generate multiple layers of an input image automatically to enable the image tester to select the highest quality detected edges. In addition, the effect of intensive I/O operations and the number of parallel running processes on the performance of the proposed techniques have also been studied.