针对已有连通域标记算法不能适用于遥感数据等海量灰度图像八连通域快速标记的问题,设计并实现了一种MGSI-8CA标记算法。该算法对传统的八连通域标记算法进行八向变四向优化,采用列表机制解决连通域标记冲突问题,并引入了筛选机制和分块处理机制,实现了对海量灰度图像八连通域标记的快速处理,同时也得到了灰度图像连通域的正确划分及精确的连通域数目。实验结果证明,MGSI-8CA标记算法运行效率高,更能适应海量灰度图像处理的要求。
To solve the problem that the existing labeling algorithm of connecting area can not be applicable for fast labeling in 8adjacent connecting area for massive gray images, such as remote sensing data, etc, this paper designs and implements a label ing algorithm of 8adjacent connecting area for massive gray scale images. Through finding from four directions of every pixel instead of from eight directions in traditional labeling algorithm, it employs the list mechanism to solve labeling conflict of con necting area, and introduces the screening mechanism and the block processing mechanism, the algorithm realizes fast process ing in 8adjacent connecting area labeling for massive gray scale images. At the same time, the algorithm also gets a correct divi sion in the connecting area of gray scale images and the accurate number of connecting area. Experimental result shows that the algorithm has high efficiency, and it is more adaptable to fulfill the requirements of massive gray scale images.