| 融合对称特性的混合标签传递半监督直觉模糊聚类图像分割 |
| |
| 引用本文： | 赵凤,吝晓娟,刘汉强.融合对称特性的混合标签传递半监督直觉模糊聚类图像分割[J].信号处理,2020,36(9):1544-1556. |
| |
| 作者姓名： | 赵凤 吝晓娟 刘汉强 |
| |
| 作者单位： | 西安邮电大学通信与信息工程学院(人工智能学院) |
| |
| 基金项目： | 国家自然科学基金(61571361, 61671377,61102095)；西安邮电大学西邮新星团队基金(xyt2016-01) |
| |
| 摘 要： | 现有的直觉模糊聚类算法应用于图像分割时，往往只考虑图像的像素信息，忽略了图像的几何特征和区域信息，使得分割效果不太理想。为了提高直觉模糊聚类算法的分割性能，提出一种融合对称特性的混合标签传递半监督直觉模糊聚类算法。该算法首先对图像进行对称轴检测获取图像的对称特性，接着利用图像的对称特性进行对称像素的标签传递并改进像素对聚类中心的直觉模糊距离测度，然后设计一种混合标签传递半监督策略，对所有像素进行隶属度的估计并将其作为监督隶属度进行引入，随后构建融合对称特性的混合标签传递半监督直觉模糊聚类目标函数，通过聚类获得最终的分割结果。两个彩色图像库上的实验结果表明，该算法能够将目标从复杂背景中完整的分割出来，分割性能优于对比算法。
|
| 关 键 词： | 图像分割 直觉模糊聚类 半监督聚类 对称特性 混合标签传递 |
| 收稿时间： | 2020-06-16 |
Hybrid Label Propagation Semi-supervised Intuitionistic Fuzzy Clustering incorporating Symmetric Property for Image Segmentation |
| |
| Affiliation: | School of Communication and Information Engineering & School of Artificial Intelligence, Xi’an University of Posts and TelecommunicationsKey Laboratory of Electronic Information Application Technology for Scene Investigation of Ministry of Public Security, Xi’an University of Posts and Telecommunications |
| |
| Abstract: | When the existing intuitionistic fuzzy clustering algorithms are applied to image segmentation, only the pixel information of the image is considered. Furthermore, due to neglecting the geometric features and regional information of the image, the segmentation result is not ideal. In order to boost the segmentation performance of the intuitionistic fuzzy clustering algorithm, a hybrid label propagation semi-supervised intuitionistic fuzzy clustering incorporating symmetric property for image segmentation algorithm was proposed. First, the symmetry axis of the image was detected to obtain the symmetric property. Second, it used the symmetry characteristics of the image to perform symmetric label propagation of pixels and to improve the intuitionistic fuzzy distance measure between the pixels and cluster centers. Third, a hybrid label propagation semi-supervised strategy was designed to estimate the membership of all pixels. It introduced the estimated membership as the supervised membership into the intuitionistic fuzzy clustering algorithm. Fourth, it constructed a hybrid label propagation semi-supervised intuitionistic fuzzy clustering objective function incorporating symmetric property. Finally, the final segmentation result was gained by the proposed algorithm. Experimental results on two color image libraries demonstrated that the proposed algorithm could segment the target from the complex background completely, and the segmentation performance was superior to the comparison algorithms. |
| |
| Keywords: | |
|
| 点击此处可从《信号处理》浏览原始摘要信息 |
|
点击此处可从《信号处理》下载全文 |
