欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
聚类算法在对图像进行分割的过程中要面对如何自动确定聚类类别数、如何克服图像特征点分布复杂的流形结构、如何减少算法的运行时间。针对这些问题，提出了流形距离的自动免疫克隆聚类图像分割算法。自动免疫克隆聚类算法可以自动确定聚类个数，不需要人为事先给定，并且确保全局收敛；使用流形距离可以反映空间分布复杂的流形数据；使用超像素而非像素来降低图像分割的时间等问题。通过对4组人工数据集和4幅自然图像进行实验，对比k-means算法、GCUK算法，结果表明该方法优势比较明显，具有一定的实用性和先进性。
英文摘要：
There are several difficulties in using a partitional clustering algorithm to deal with image segmentation problem including choosing the correct number of clusters without any prior knowledge, measuring the image datasets with complicated manifold structures and reducing the computation time. In this paper, an automatic immune clonal clustering method using manifold distance is applied to image segmentation. This method can automatically determine the number of clusters, measure the complicated manifold dataset by using manifold distance, and less computation time by using super-pixels instead of pixels. Experimental results on four artificial data sets and four Berkeley images show that the novel method outperforms the k-means algorithm and the GCUK algorithm.
同期刊论文项目
同项目期刊论文
Globally stable adaptive robust tracking control using RBF neural networks as feedforward compensato
Classification of aurora series image based on elgenvalue-scaling kernel fisher discriminant analysi
A Novel Immune Optimization Algorithm for Fairness Resource Allocation in Cognitive Wireless Network
A modified objective function method with feasible-guiding strategy to solve constrained multi-objec
Multi-objective evolutionary for synthetic aperture radar image segmentation with non-local means de
期刊信息
位置：