一种新的熵的提取方法在图像分割中的应用

针对正态云的正向云模型中熵的求法存在着误差大、计算量大的问题，提出了一种基于云理论的“3E”规则的熵的求法．首先将图像灰度信息作出统计，将波峰值作为Ex，再将距离波峰最近的第一个波谷值与波峰之间的距离作为3En最后通过一定规则微调得到符合条件的En.并通过实验将云模型与阈值法分割图像相结合，实验结果证明，该方法不但可以准确计算出己，使误差减少、计算量也较小，并且对图像分割效果好、效率高．

Application of a new entropy extraction method in image segmentation

To overcome the large error and heavy computation existed in solving entropy for the normal, positive cloud model, an entropy extraction method for ＂3En＂ rule based on cloud theory is proposed in this paper. First, the statistic information on the gray level of an image is given, in which Ex is the wave crest, 3En is the distance between the wave crest and first trough closest to the wave crest. At last by free-tuning based on a certain rule, the eligible E, that accords with the condition is obtained. And through experiment the cloud model is combined with the image segmented by threshold method. The experimental result proves that this method can accurately calculate the En, which can reduce the error and computation amount, segment the image with better effect and high efficiency.