An Algorithm Based On Distance For Binary Image Labeling

With the progress of image processing technology,in pattern recognition,testing technology,etc,binary image has been a fundamental problem of research.Binary image is simple,takes up less space and fast processing speed.Binary image labeling is a very important step in image analysis,especially in binary connected components labeling.It has become one of the basic problems in the field of image processing.In this paper,the algorithm is proposed of binary image labeling method based on distance.The distance is used to solve such a problem.The nature of labeling is more clear.This paper first introduces the traditional methods.A common shortcoming of the traditional methods is that the binary image is needed to scan repeatly.In the method based on scanning pixel lines,a connected component contains different sub connected components labeling.The merge analysis is needed.The region growing method is fit for labeling a small area.When the area is too large,the buffer can not accommodate or the efficiency is low.For the eight neighborhood of each pixel are needed to repeatedly check.Breadth-first-search based on region growing avoid the problem by using iterative thought instead of recursive thoughts.In this paper,the binary image labeling method based on distance is presented.The labeling problem can be solved one-time and improve the efficiency for the point dispersion image.The algorithm not only can label the whole area one-time,but also repeated labeling is avoided.It is suitable for more types of images and has good robustness.