Pixel scratching algorithm for edge detection of monochrome image objects

摘要

In this paper, we propose a simple and new technique called as 'pixel scratching algorithm'. This method finds edges of objects in any black and white image by scratching out all those pixels that show abrupt changes in grey levels and making all other pixels as background. These scratched out pixels then form a sketch of the edge of the objects that are present in the image. In contrast to existing edge detection techniques like Laplacian operators, Sobel operators, Prewitt operators, the proposed technique gives finer edges with much less computation without the need of any mask operator, finding second derivatives or applying weight for smoothing. In this approach, edge pixels are determined by scanning the image line by line from top to bottom and finding out differences between successive pixel values, followed by thresholding. Experimental results show that the proposed method provides more promising result than the methods being compared here.