地球信息科学学报

地球信息科学学报 ›› 2014, Vol. 16 ›› Issue (1): 142-150.doi: 10.3724/SP.J.1047.2014.00142

• 遥感科学与应用技术 • 上一篇    

改进的分水岭变换算法在高分辨率遥感影像多尺度分割中的应用

张博, 何彬彬   

  1. 电子科技大学资源与环境学院, 成都 611731
  • 收稿日期:2013-04-02 修回日期:2013-05-02 出版日期:2014-01-05 发布日期:2014-01-05
  • 通讯作者: 何彬彬（1972-），男，教授，博导，主要研究方向为定量遥感、空间数据挖掘、遥感图像处理与识别。E-mail：binbinhe@uestc.edu.cn E-mail:binbinhe@uestc.edu.cn
  • 作者简介:张博（1988-），男，硕士生，研究方向为遥感图像处理。E-mail：zhangbomoon@gmail.com
  • 基金资助:

    国家自然科学基金项目“耦合不确定性空间推理和案例推理的区域矿产资源潜力预测模型研究”（41171302）。

Multi-scale Segmentation of High-resolution Remote Sensing Image Based on Improved Watershed Transformation

ZHANG Bo, HE Binbin   

  1. School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China
  • Received:2013-04-02 Revised:2013-05-02 Online:2014-01-05 Published:2014-01-05

PDF (PC)

赞

被引次数

16

摘要：

由于高空间分辨率遥感影像自身的复杂性，传统的分水岭分割方法难以取得令人满意的效果。本文提出一种改进分水岭变换的高分辨率遥感影像多尺度分割方法，在抑制分水岭过分割现象的同时，还能实现对遥感影像的多尺度分割。该方法充分考虑了高分辨率遥感影像的多光谱和多尺度特性，首先，利用各向异性扩散滤波技术对影像进行平滑滤波，目的是在滤除各种噪声的同时还能保持影像的边缘特征和重要的细节信息；然后，提取影像的多尺度形态学梯度，并从梯度图像中提取标记；接着进行基于标记的分水岭变换；最后，利用改进的快速区域合并算法实现对影像的多尺度分割。实验表明，改进的算法能有效地抑制分水岭的过分割现象，对高分辨率遥感影像有较好的分割性能。

关键词: 影像分割, 高分辨率遥感影像, 多尺度, 分水岭

Abstract:

With the development of high resolution remote sensing images, imaging analysis technology of object-oriented method shows a distinct advantage in the field of information extraction and target recognition. Image segmentation, as a key technology of object-oriented image analysis method, has a vital role to play on the latter feature extraction and application analysis. Watershed transformation is usually adopted for image segmentation because of its unique advantages. However, because of the complexities of high spatial resolution remote sensing image itself, the traditional method of watershed segmentation is difficult to obtain satisfactory results. This paper presents a new multi-scale segmentation method for high resolution remote sensing image based on improved watershed transformation, in order to suppress over-segmentation of watershed transformation, as well as to provide arbitrary-scale segmentation of remote sensing image for object-oriented segmentation method. The algorithm fully considered multi-spectrum, multi-scale and multi-noises characteristics of high spatial resolution remote sensing image. The details are described as follows. Firstly, an anisotropic diffusion filter was used for image smoothing, because this technology can both remove the noises and maintain edges and other important details information of the input image. Secondly, in order to take into account the multi-scale characteristics of remote sensing images, multi-scale morphology gradient was extracted because of its good combination of the advantages of large structural element and small structural element, and then H-minima technology was used to extract tags of gradient image for the latter marker-based watershed algorithm. Finally, an improved fast region-merging algorithm was proposed to achieve the multi-scale segmentation. This paper elaborated the pre-processing filtering, multi-scale gradient, marking extraction and multi-scale region merging aspects, and the experiments showed that the proposed segmentation method effectively restrained over-segmentation of watershed transformation, and had a good performance for segmentation of high spatial resolution remote sensing images.

Key words: high-resolution remote sensing image, image segmentation, watershed, multi-scale