Description: 一种 较新的聚类算法 Dominant-set 的代码，包括聚类算法的代码和测试代码。该算法最大特点 就是基于图理论的 ，相对于Normalized Cut,计算复杂度低很多，况且能自动决定类的个数 -A relatively new clustering algorithm Dominant-set the code, including the clustering algorithm code and test code. Most prominent feature of the algorithm is based on graph theory, and compared with the Normalized Cut, much lower computational complexity, decision Moreover automatically the number of categories Platform: |
Size: 3072 |
Author:曾祥林 |
Hits:
Description: 使用聚类中K－平均算法，以颜色分量作为坐标参数，对景象图进行聚类分析，要求最后的分类结果将路标（可能包括少量相似区域）聚类为一个模式类别-The use of clustering in the K-average algorithm, to the color component parameters as the coordinates of a scene graph for cluster analysis requires the classification of the final results will be signs (which may include a small number of similar regional) clustering as a model category Platform: |
Size: 1956864 |
Author:hddnudt |
Hits:
Description: 针对大规模稀疏无向图开发的聚类算法。附带测试页面-this program is supposed to embark on the undiret graph clustering issue. Platform: |
Size: 141312 |
Author:zhangxucheng |
Hits:
Description: 在左视图上单击鼠标左键，可获得3种数据源：【标准数字聚类】、【手画图形聚类】、【位图文件分析聚类】。
(1) 标准数字
在工具条中按下【标准数字聚类】按钮后，选择工具条上提供的各种标准数字。在左视图就会得到多个标准数字。
每行中存放的标准数字个数与blank.bmp文件大小有关，读者可以自行修改该文件的大小，应注意该文件应该是n×n的，比如500×500 。
(2)手写数字
在工具条中按下【手画图形聚类】按钮后，拖动鼠标左键画各种数字或图形，注意每一个物体要连通。
(3) 打开位图文件
在工具条中按下【位图文件分析聚类】按钮后，打开需要聚类分析的位图文件。 弹出“打开文件”对话框，读者可以打开已经存在的一幅图像文件。
-In the left view, click the left mouse button, three kinds of data sources available: standard digital cluster 【】, 【hand-painted graphics cluster】, 【】 clustering analysis of bitmap file.
(1) Standard Digital
In the tool bar by pressing 【】 button standard digital cluster, select the toolbar offers a variety of standard digital. In the left view will be more than standard digital.
Each line number stored in standard digital file with the size of blank.bmp, readers can modify the size of the document should be noted that the document should be n × n, such as 500 × 500.
(2) handwritten numeral
Click in the tool bar graph clustering】 【hand-painted button, drag the left mouse button draw a variety of numbers, or graphics, attention to every object to be connected.
(3) Open the bitmap file
In the toolbar bitmap file by pressing 【】 button cluster analysis, cluster analysis of the need to open the bitmap file. Pop-up "Open File" dialog box, the reader can open an existing image fi Platform: |
Size: 6771712 |
Author:哈哈 |
Hits:
Description: In recent years, spectral clustering has become one of the most popular modern clustering algorithms.
It is simple to implement, can be solved efficiently by standard linear algebra software, and very often outperforms
traditional clustering algorithms such as the k-means algorithm. Nevertheless, on the first glance spectral clustering
looks a bit mysterious, and it is not obvious to see why it works at all and what it really does. This article is a
tutorial introduction to spectral clustering. We describe different graph Laplacians and their basic properties,
present the most common spectral clustering algorithms, and derive those algorithms from scratch by several
different approaches. Advantages and disadvantages of the different spectral clustering algorithms are discussed. Platform: |
Size: 353280 |
Author:cc |
Hits:
Description: Weintroduceanewsimilaritymeasurebetweendatapointssuited
respecttoclustercentroids),shareswiththepreviousapproachthe
for clustering and classi?cation on smooth manifolds. The pro-
sameproblemsin itsoptimization formulation.
posed measure is constructed from a dual rooted graph diffusion
Recently, the focus of attention in unsupervised learning has
over the feature vector space, obtained by growing dual rooted
turned to spectral clustering methods due to its many successes
minimum spanning trees (MST) between data points. This diffu-
[1]. These methods use the spectral content of a similarity ma-
sionmodelforpairwiseaf?nitiesnaturallyac Platform: |
Size: 204800 |
Author:quinquindavid |
Hits:
Description: 提出一种结合小波变换与共现矩阵用于纺织品图像缺陷检测的方法。首先将灰度图像分解成子带 然
后将纹理图像分割成互不重叠的子窗口, 提取共现特征 最后用无缺陷样品训练的M ahalanob is分类器将每一子
窗口划分为缺陷的和无缺陷的。应用该算法进行实际工厂环境中的纺织品缺陷检测。实验结果表明, 集中处理
具有强判决能力的某一频带提高了检测性能, 也改善了计算效率。-Propose a wavelet transform and co-occurrence matrix for the textile image defect detection method. First, the gray image is decomposed into sub-band and then the texture image into non-overlapping sub-windows were now feature extraction Finally, defect-free samples of M ahalanob is trained classifier to each child window is divided into defective and non- defects. Practical application of the algorithm textile factory defect detection in the environment. Experimental results show that the decision to focus with a strong ability to improve the detection performance of a band, but also improve the computation efficiency. Platform: |
Size: 229376 |
Author:胡 |
Hits:
Description: 内容是英文的，是关于张量在图聚类上的应用的-Content is in English, is the tensor in the graph on the application of clustering Platform: |
Size: 201728 |
Author:lyn |
Hits:
Search - graph clustering