摘要基于图割的交互式图像分割方法以其简单的交互方式，较好的分割结果等特性获得了广泛的关注。对于背景简单的图像，用户只需提供较少的种子点，图割算法就能自动获取满意的分割结果。然而，对于噪声以及纹理图像，传统的图割算法往往较难获得满意的结果。为了克服以上问题，进一步提升图割算法的分割精度，本文在传统图割模型基础上有效地引入图像片级信息，通过比较两个图像片间的相似性来获得更加精确的特征。已有的基于图像片的分割方法一般将图像片级信息引入能量函数的边界项，由于需要计算所有图像片间的相似性，极大地增加了算法的时间复杂度。为了在图割模型中有效地利用图像片信息，本文首先使用混合高斯模型（GMM）来为图像片的结构特征建模，这样基于GMM的学习，我们可以获得图像片级信息。然后将像素级信息与图像片级信息相结合，以进一步提高边界区域细节的分割精度。本文最后对所提出的算法进行了大量实验，验证了其有效性和鲁棒性。28217
关键词  图像分割  非局部  图像片  图割
毕业论文设计说明书外文摘要
Title   The research on combining adaptive patch and the interactive  image segmentation based on graph cut  
Abstract
The interactive image segmentation algorithm based on graph cut has attracted considerable attention due to its simple interaction, better segmentation results and other features. For image with simple background, user only need to provide less seeds and then the graph cut algorithm can automatically obtain a satisfactory segmentation result. However it always obtains a satisfactory result difficultly for image with noise and texture. In order to overcome above problems and further improve the accuracy of the segmentation, in this paper, patch-level information is effective introduced into the traditional graph cut model, in which the more accurate feature is utilized by comparing the similarities between two patches. The segmentation methods based on patches usually introduce the patch-level information into the boundary term of the energy function, which would dramatically increase the time complexity due to the similarity computations of all patches. To effectively utilize the patch information in graph cut model, in this paper, we first utilize the Gaussian mixture model (GMM) to construct the structural features of patches. Therefore, the patch-level information can be obtained based on the learning of GMM. Then we combine the pixel-level information and patch-level information together to further improve the segmentation accuracy for the details around boundary regions. The performance of our proposed approach demonstrates its improved robustness and effectiveness.
Keywords  Image segmentation  Nonlocal  Image patch  Graph cut
目   次
1  绪论  1
1.1    图像分割研究背景  1
1.2    交互式图像分割研究现状  2
1.3    本文主要内容及章节安排  4
2      图像分割的理论基础  5
2.1  图像分割的定义  5
2.2  图像分割方法的分类  6
2.3  分割特征及描述方法  6
2.4  本章小结  9
3  基于图割理论的图像分割 10
3.1  图理论  10
3.2  s-t网络流图与其割  11
3.3  基于图割理论的图像分割  12
3.4  基于非局部方法的模型  14
3.5  本章小结  15
4  结合像素和图像片的图像分割算法  16
4.1  高斯混合模型  16
4.2  本文算法  17
4.3  实验结果与分析  21
4.4  本章小结  25 结合自适应图像片与图割的交互式图像分割算法研究:http://www.751com.cn/jisuanji/lunwen_22982.html