二维最大相关准则阈值分割的快速算法  

A fast image thresholding method based on two-dimensional maximum correlation criterion

作　　者：何龙[1] 赵勇[2] 

机构地区：[1]成都航空职业技术学院机械工程系,四川成都610021 [2]西北工业大学机电学院,陕西西安710072

出　　处：《电子技术应用》2008年第7期32-34,共3页Application of Electronic Technique

摘　　要：提出了一种快速二维最大相关准则阈值分割算法。该方法利用积分和迭代编程技术,避免了数据的重复运算,从而将计算的复杂性从一般二维最大相关准则方法的O(L^4)减少至O(L^2)。实验结果表明,该方法分割效果好,计算效率高,适合实时图像系统的应用。A fast image thresholding method based on two-dimensional maximum correlation criterion（MCC） is proposed. The method uses an integral and recursive programming technique which reduces a mass of repeated data computation, therefore the computation complexity for the proposed method is reduced from O（L^4）to O（L^2）. The experimental results show that the proposed method can get ideal segmentation result and is suited for real-time applications.

关 键 词：阈值分割 最大相关准则 快速算法 

分 类 号：TN911.73[电子电信—通信与信息系统] TP391.41[电子电信—信息与通信工程]