摘要： 为降低噪声对图像分割结果的影响，提出一种基于热平衡理论的中智学图像分割方法。该方法将图像转化为中智学图像，考虑每一个像素的不确定性，通过热平衡运算及图像增强处理，使噪声点变得更平滑，再使用γ-均值聚类方法分割图像。实验结果表明，对于含不同程度噪声的图像，该方法的分割效果明显优于中智学方法及改进的模糊C-均值方法。

关键词: 图像分割, 中智学, 热平衡, 图像增强, 聚类

Abstract: In order to reduce the influence on image segmentation effect caused by noise, this paper proposes a neutrosophic image segmentation approach based on thermal balance theory. After transforming the image to neutrosophic image, by considering the indeterminacy of each pixel, it smooth each noise pixel through thermal balance and enhancement operations. And the image is segmented by using γ-Means clustering method. Experimental results show that the method can obtain better segmentation effects than neutrosophy method and Modified Fuzzy C-Means(MFCM) method for images with different degree noise.

Key words: image segmentation, neutrosophy, thermal balance, image enhancement, clustering

中图分类号: 

• TN911.73