摘要
熵阈值法是图像分割中的重要方法,并在图像处理中得到了广泛的应用。针对香农熵阈值法因存在对数计算而导致计算量过大的问题,本文首先提出了一种新的信息熵;其次对其一些性质进行了探讨;最后用于图像分割的阈值选取。实验结果表明,本文的新熵阈值法是可行的,且其计算所需时间量远小于比香农熵阈值法。
Entropy based thresholding is an important method of image segmentation and is extensively used in the image processing for many applications. This paper considers that thresholding based on Shannon entropy has some shortage of more computation time due to logarithm operation. The new information entropy is firstly proposed in the paper, then some of its properties are discussed. Finally it is used to choose optimal threshold in image segmentation. The experimental results show that new entropy based thresholding method is feasibile, and its cost time is far less than that of thresholding method based on Shannon entropy.
出处
《西安邮电学院学报》
2009年第1期72-79,共8页
Journal of Xi'an Institute of Posts and Telecommunications
基金
国家自然科学基金项目(编号:60572133)
陕西省教育厅项目(编号:06JK194)
