Description: 有向无环图支持向量(DAG-SVMS)多类分类方法,是一种新的多类分类方法。该方法采用了最小超球体类包含作为层次分类依据。试验结果表明,采用该方法进行多类分类,跟已有的分类方法相比有更高的分类精度。
-Directed acyclic graph support vector (DAG-SVMS) multi-category classification methods, is a new multi-category classification methods. The method uses the smallest category of super-sphere that contains the level of classification as a basis. The experimental results show that using the method of multiclass classification with the classification method has been compared to a higher classification accuracy. Platform: |
Size: 5120 |
Author:苏苏 |
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: 将图像中的物体按圆形度分类，并以不同色彩的边界标示圆形、矩形、不规则的阔图形、狭长图形4类图斑-Images of objects according to a circular classification, and the borders of different colors marked round, rectangular, irregular width graphics, strip chart graph 4 spot Platform: |
Size: 1024 |
Author:朱燕玲 |
Hits:
Description: 1．系统需要为不同的用户设立不同的帐户。用户在创建新账号时候需要设定用户名和密码。使用时，用户需要输入密码才能进入自己的帐户；
2．用户可以添加、修改和删除收入和支出信息，信息包括：交易日期、收支分类、币种、金额以及说明等，系统同时显示帐户余额；
3．用户可以根据实践范围和币种查询收支信息；
4．系统可以按用户设定的实践段统计日常收支，以曲线图显示逐月变化情况
-1. Systems required for different users to set up a different account. Users at the time required to create a new account set up user name and password. Use, users need to enter a password to access their own accounts 2. The user can add, modify, and delete the income and expenditure information, including information: transaction date, revenue and expenditure classification, currency, amount and description, the system shows account balance at the same time 3. Users can query the practice of the scope and currency income and expenditure information 4. System can be set by user day-to-day practice of balance of payments statistics paragraph to graph shows the monthly Change Platform: |
Size: 16384 |
Author:王小天 |
Hits:
Description: 基于弹性模板匹配的人脸表情识别程序。首先针对静态表情图像进行表情图像的灰度、尺寸归一化，然后利用Gabor小波变换提取人脸表情特征以构造表情弹性图，最后提出基于弹性模板匹配及K-近邻的分类算法实现人脸表情的识别。-Flexible template matching based on facial expression recognition procedures. First of all, the expression for the static image of the gray-scale face image, size, normalized, and then extracted using Gabor wavelet transform features of human facial expression to expression of elastic graph structure, and finally based on flexible template matching and the K-neighbors classification algorithm Facial Expression identification. Platform: |
Size: 763904 |
Author:hejian |
Hits:
Description: For now, we are not interested in what this graph represents. Rather, we would like to "save" the classification results from which the graph is generated. In the new window, we click on the "Save" button and save the result as the file: "bank-predicted.arff", Platform: |
Size: 8192 |
Author:poorni |
Hits:
Description: 内容管理
分类部分支持无级分类，可以对每个分类指定显示模式，单页模式（指为独立的一个网页，如公司简介，联系我们），列表模式（一列一列的显示），展示模式（主要用于产品展示使用，列表页主要体现为一张小图一个标题（或者为简介，此处可人为定义））；
图片|视频：此功能主要是为当前单页或分类定义一张图片或FLASH等；
关联功能：主要用于分类或单页面之间的关联，如：加了一个单页：关于我们，现在想在关于我们中左侧显示联系我们，这时我们就可以使用些功能来选择了；
内容管理中支持三种信息模式，普通、图片|视频（主要用于每个不同的页面上方可以选择不同的表现图片或FLASH动画，同时支持FLV流媒体）、URL（点击信息标题直接转到指定的URL中）；
信息属性定义，推荐、固顶、热点等；
超强的编辑器
-Content Management
Category partial support for non-class categories, you can specify the display mode for each category, single-page mode (refer to separate a Web page, such as About Us, Contact Us), list mode (an a display), display mode (mainly for Products to use, the list page is mainly reflected as a small graph a title (or for us, here artificially defined))
Photo | Video: This feature is mainly for the current definition of a single page or a picture, or FLASH classification, etc.
Correlation function: mainly used for classification or association between a single page, such as: adding a single page: About us, and now would like to display on the left of us, contact us, then we can use some functions to select a
Content management information in support of three modes, Normal, Photo | videos (mainly top of the page for each different performance can choose different pictures or FLASH animation, also supports FLV streaming media), URL (click the title to go directly t Platform: |
Size: 3190784 |
Author:yangyi |
Hits:
Description: 1.GeometricContext文件是完成图片中几何方向目标分类。
参考文献《Automatic Photo Pop-up》Hoiem 2005
2 GrabCut文件是完成图像中目标交互式分割
参考文献《“GrabCut” — Interactive Foreground Extraction using Iterated Graph Cuts》
C. Rother 2004
3 HOG文件是自己编写的根据HOG特征检测行人的matlab代码
4 虹膜识别程序是下载的一个通用的虹膜识别程序，可以运行
5 GML_AdaBoost_Matlab_Toolbox是一个很好用的adaboost matlab工具箱
6 libsvm-mat-2.91-1 是用C编写的改进的SVM程序，代码质量很高，提供了matlab接口
7 SIFT_Matlab 是编写的利用sift特征进行的宽基线匹配，代码质量高
8 FLDfisher 是利用fisher 线性降维方法进行人脸识别-1.GeometricContext file is complete the picture in the geometric direction of target classification. References " Automatic Photo Pop-up" Hoiem 2005 2 GrabCut the target file is an interactive segmentation of image reference " " GrabCut " - Interactive Foreground Extraction using Iterated Graph Cuts" C. Rother 2004 3 HOG documents prepared under their own HOG Characteristics of pedestrian detection matlab code 4 iris recognition process is to download a general iris recognition program, you can run 5 GML_AdaBoost_Matlab_Toolbox is a good use of adaboost matlab toolbox 6 libsvm-mat-2.91-1 is written in C to improve the SVM procedures, code of high quality, provides a matlab interface to 7 SIFT_Matlab is prepared for the use of sift features a wide baseline matching, the code is the use of high quality 8 FLDfisher fisher linear dimension reduction method for face recognition Platform: |
Size: 6918144 |
Author:张数 |
Hits:
Description: RSE (Regularized Selective Ensemble) is a selective ensemble learning algorithm for binay classification, which constructs ensemble under the regularization framework. In current version, the graph Laplacian serves as the regularizer, and unlabeled data can also be exploited to improve the performance.
Platform: |
Size: 23552 |
Author:123 |
Hits:
Description: 有向无环图的多类支持向量机分类算法A directed acyclic graph multi class classification algorithm of support vector machine-A directed acyclic graph multi class classification algorithm of support vector machine Platform: |
Size: 139264 |
Author: |
Hits:
Description: 关于metric learning的综述，涉及到许多的知识：SVM、kernel、SDP等-This paper surveys the field of distance
metric learning from a principle perspective, and includes a broad selection of recent work. In particular, distance metric learning is reviewed under different
learning conditions: supervised learning versus unsupervised learning, learning in a global sense versus in a local sense and the distance matrix based on linear kernel versus nonlinear kernel. In addition, this paper discusses a number of techniques
that is central to distance metric learning, including convex programming, positive semi-definite programming, kernel learning, dimension reduction, K Nearest Neighbor, large margin classification, and graph-based approaches. Platform: |
Size: 322560 |
Author:刘建飞 |
Hits:
Description: 在图像识别技术的实现过程中，图像分割是一个重要的预处理环节，图像分割效果，直接影响着后续的分类、目标识别、图像分析、图像理解等过程的结果。针对着不同的图像特点，目前已经提出了错综复杂的图像分割算法。其中基于图论的图像分割算法是近几年研究的热点，这类算法着眼于全局，更注重局部数据的处理，比一般方法可以获得更佳的效果，并且图论理论有着比较完备的数学理论基础，将其用于图像处理有着较好的应用前景。-In the implementation process of image recognition technology, image segmentation is an important pre-processing aspects of image segmentation directly affect the subsequent classification, target recognition, image analysis, image understanding the results of the process. For different image characteristics, the intricate image segmentation algorithm has been proposed. The hot image segmentation algorithm based on graph theory research in recent years, this type of algorithm looks at the overall situation and focus more on the local data processing can get better results than the average, and graph theory, the theory has a more complete mathematical theory the basis for image processing has good application prospects. Platform: |
Size: 3821568 |
Author:xingzhiwen |
Hits:
Description: 图像分割是一种重要的图像分析技术，对图像分割的研究一直是图像技术研究中的热点和焦点。分水岭算法是基于数学形态学理论的图像分割算法，但是对噪声敏感且存在过分割的现象。为提高图像分割效果，本文通过对影像进行滤波处理，用改进的快速区域合并算法优化分水岭算法进行影像分割。-In this work a technique for cloud detection and classification from MSG–SEVIRI (Meteosat Second Generation–Spinning Enhanced
Visible and Infra-red Imager) imagery is presented. It is based on the segmentation of the multispectral images using order-invariant
watershed algorithms, which are applied to the corresponding gradient images, computed by a multi-dimensional morphological operator.
To reduce the over-segmentation produced by the watershed method, a RAG (Region Adjacency Graph) based region merging
technique is applied, using region dissimilarity functions. Platform: |
Size: 2011136 |
Author:comewanlei |
Hits:
Description: Hyperspectral image classification with limited
number of labeled pixels is a challenging task. In this paper, we
propose a bilayer graph-based learning framework to address
this problem. For graph-based classification, how to establish
the neighboring relationship among the pixels the high
dimensional features is the key toward a successful classification.
Our graph learning algorithm contains two layers. The first-layer
constructs a simple graph, where each vertex denotes one pixel
and the edge weight encodes the similarity between two pixels.
Unsupervised learning is then conducted to estimate the grouping
relations among different pixels. These relations are subsequently
fed into the second layer to form a hypergraph structure, on top
of which, semisupervised transductive learning is conducted to
obtain the final classification results. Our experiments on three
data sets demonstrate the merits of our proposed approach,
which compares favorably with state of the art.-Hyperspectral image classification with limited
number of labeled pixels is a challenging task. In this paper, we
propose a bilayer graph-based learning framework to address
this problem. For graph-based classification, how to establish
the neighboring relationship among the pixels the high
dimensional features is the key toward a successful classification.
Our graph learning algorithm contains two layers. The first-layer
constructs a simple graph, where each vertex denotes one pixel
and the edge weight encodes the similarity between two pixels.
Unsupervised learning is then conducted to estimate the grouping
relations among different pixels. These relations are subsequently
fed into the second layer to form a hypergraph structure, on top
of which, semisupervised transductive learning is conducted to
obtain the final classification results. Our experiments on three
data sets demonstrate the merits of our proposed approach,
which compares favorably with state of the art. Platform: |
Size: 2847744 |
Author:bala |
Hits:
Search - Graph Classification