摘要： 最大类间方差(Otsu)在用于含噪图像的分割时，不能取得理想的分割效果。针对该问题，在Otsu分割方法的基础上，结合小波包分析，提出一种新的含噪遥感图像分割算法。该算法用小波包对含噪遥感图像进行分解，采用自适应阈值的方法去除图像中的噪声，应用Otsu方法对图像进行分割。仿真实验表明，该算法具有较好的抗噪性能及分割效果。

关键词: 遥感图像, 阈值分割, 小波包变化, 自适应阈值, 最大类间方差分割

Abstract: When the variance between the largest category(Otsu) is used in noise image, Otsu method can not provide satisfactory results. Under this situation, this paper proposes a new noisy remote sense image segmentation method. The method applies wavelet package transform to decompose the remote image. The noises in the remote image are eliminated by the adaptive threshold method. Otsu method is adopted to segment image. Simulation experiments show that the algorithm has better immunity to noise and achieves good performance in segmentation quality.

Key words: remote sense image, threshold segmentation, wavelet packet transformation, adaptive threshold, Otsu segmentation

中图分类号: 

• TP391