ENTROPIC THRESHOLDING
ENTROPIC THRESHOLDING
复制标题
DOI：
10.1016/0165-1684(89)90090-x
复制
发表时间：
1989-02-01
影响因子：
4.4
通讯作者：
PAL, SK
中科院分区：
文献类型：
--
作者：
PAL, NR;PAL, SK
定义了图像的有序熵(取决于长度为q的灰度级序列中包含的信息)和图像的条件熵。发现当q=1时，现有的全局熵定义是它的特例。利用这些定义，提出了两种图像阈值分割算法(目标-背景分类)，并利用其共生矩阵实现了该算法。对于一组图像，实验证明了这些算法的优越性。
Entropy of orderq(depending on the information contained in a sequence of gray levels of lengthq) and the conditional entropy of an image are defined. The existing definition of global entropy is found to be its special case whenq= 1.Using these definitions, two algorithms for image thresholding (object-background classification) are formulated and implemented with the help of its co-occurrence matrix. The superiority of these algorithms is experimentally demonstrated for a set of images.