Description: 利用聚类算法分割图像，将维对分法只可将图像分为2部分，可以作为二值化的代码,K-均值法可将图像分为任意多部分。程序直接采用R、G、B三色作为特征参数，聚类中心为随机值，当然也可以采用其他参数，程序编译为EXE文件后速度还可以接受，但尚有改进的余地，那位高手有空修改的话，请给我也发份代码。-clustering algorithm using image segmentation, Victoria right method can only image is divided into two parts, the two values can be used as the source, K-means algorithm can be divided into images of arbitrary multi-part. Procedures used directly in R, G, B color as the characteristic parameters for the cluster center random value, of course, can also be used for other parameters, procedures EXE compiler to speed document acceptable, but there is still room for improvement, but the master of the time change, then please give me also made in the code. Platform: |
Size: 50271 |
Author:pbt |
Hits:
Description: 利用聚类算法分割图像，将维对分法只可将图像分为2部分，可以作为二值化的代码,K-均值法可将图像分为任意多部分。程序直接采用R、G、B三色作为特征参数，聚类中心为随机值，当然也可以采用其他参数，程序编译为EXE文件后速度还可以接受，但尚有改进的余地，那位高手有空修改的话，请给我也发份代码。-clustering algorithm using image segmentation, Victoria right method can only image is divided into two parts, the two values can be used as the source, K-means algorithm can be divided into images of arbitrary multi-part. Procedures used directly in R, G, B color as the characteristic parameters for the cluster center random value, of course, can also be used for other parameters, procedures EXE compiler to speed document acceptable, but there is still room for improvement, but the master of the time change, then please give me also made in the code. Platform: |
Size: 50176 |
Author:pbt |
Hits:
Description: 图像集群（Image Clustering）
(1)图像读入，显示图像所在路径；
(2)采用imgcluster函数进行图像集群，选择集群个数后进行图像集群；
(3)运行后，在原图像上显示集群灰度图；
(4)若要显示各个集群情况，可打开【Show Clustering Image】新窗体，显示各集群类的基于原图的彩绘区域。其中非当前集群范围，则显示灰度为255的黑色。用户可点击按纽上下查看所有集群图。-image cluster (Image Clustering) (1) read into the images, Images show host path; (2) use of imgcluster function for image clusters, After the number of clusters chosen for image clusters; (3) After the operation, in the original image displayed on the gray level clusters; (4) To show that the various clusters, [Show Open Clustering Image-- new windows, showed that the cluster type based on the maximum of regional painting. Clusters of non-current range, it shows that the intensity of 255 black. Users can click on View All button next cluster map. Platform: |
Size: 113664 |
Author:mecal |
Hits:
Description: 提出了一种新颖的处理图像视觉聚类问题的方法。与以往关注图像局部特征和局部连续性的方法不同，本文中的方法能够提取关于图像的全局印象。为此，我们将图像分割问题转化为图划分问题并提出了划分中的一种全局判别准则——Ncut (Normalized Cut)。Ncut不仅能够衡量不同聚类之间的相异程度，还能够衡量各聚类内部的相似程度。为求解Ncut 的最优化问题，提出了一种基于广义特征值问题的高效算法，并将此算法应用于静态图像分割，取得了良好的效果。-proposed a novel image processing visual clustering problem. Concerned with the past, image characteristics and continuity of the partial difference in the way this method can extract images on the overall impression. To this end, We will image segmentation into the map division and the division of a global criterion-- Ncut ( Normalized Cut). Ncut different not only to measure the difference between the cluster level, but also to measure the internal cluster degree of similarity. Ncut to solve the optimization problem, based on a generalized eigenvalue problem of efficient algorithms, this algorithm will be applied to the static image segmentation, and achieved good results. Platform: |
Size: 23552 |
Author:戚颖杰 |
Hits:
Description: A general technique for the recovery of signi cant
image features is presented. The technique is based on
the mean shift algorithm, a simple nonparametric pro-
cedure for estimating density gradients. Drawbacks of
the current methods (including robust clustering) are
avoided. Feature space of any nature can be processed,
and as an example, color image segmentation is dis-
cussed. The segmentation is completely autonomous,
only its class is chosen by the user. Thus, the same
program can produce a high quality edge image, or pro-
vide, by extracting all the signi cant colors, a prepro-
cessor for content-based query systems. A 512 512
color image is analyzed in less than 10 seconds on a
standard workstation. Gray level images are handled
as color images having only the lightness coordinate-A general technique for the recovery of sig ni cannot image features is presented. The techni que is based on the mean shift algorithm, a simple nonparametric pro-cedure for estimat ing density gradients. Drawbacks of the curren t methods (including robust clustering) are av oided. Feature space of any nature can be proces sed, and as an example, color image segmentation is dis-cussed. The se gmentation is completely autonomous. only its class is chosen by the user. Thus, the same program can produce a high quality edge image, or pro-vide. by extracting all the signi cannot colors, a prepro- cessor for content-based query syste ms. A 512,512 color image is analyzed in less than 10 seconds on a standard workstation. Gray 4ISR l images are handled as color images having only the lightness c Platform: |
Size: 17408 |
Author:gggg |
Hits:
Description: K均值聚类算法 由于对纹理图像使用灰度共生矩阵分割效果不明显 因此该算法使用图像频域进行处理-K-means clustering algorithm because of the texture image segmentation using the gray co-occurrence matrix effect was not obvious, therefore use the algorithm for processing images in frequency domain Platform: |
Size: 304128 |
Author:小五子 |
Hits:
Description: To identify distinguishable clusters of data in an n-dimensional pixel
value image.
Given: Samples of multi-spectral satellite images -To identify distinguishable clusters of data in an n-dimensional pixel
value image.
Given: Samples of multi-spectral satellite images Platform: |
Size: 11264 |
Author:imran |
Hits:
Description: This paper presents a clustering approach
which estimates the specific subspace and the intrinsic dimension of each class. Our approach
adapts the Gaussian mixture model framework to high-dimensional data and estimates
the parameters which best fit the data. We obtain a robust clustering method called High-
Dimensional Data Clustering (HDDC). We apply HDDC to locate objects in natural images
in a probabilistic framework. Experiments on a recently proposed database demonstrate the
effectiveness of our clustering method for category localization. Platform: |
Size: 193536 |
Author:tra ba huy |
Hits:
Description: 将K均值算法用于图像分割,输入的是彩色图像,转换为灰度图像进行分割,输出结果为灰度图像.利用灰度做为特征对每个像素进行聚类,由于光照等原因,有时应该属于一个物体的像素,其灰度值也会有很大的差别,可能导致对该像素的聚类发生错误.在分割结果中,该物体表面会出现一些不同于其它像素的噪声点,因此,算法的最后,对结果进行一次中值滤波,以消除噪声,达到平滑图像的作用-The K means algorithm for image segmentation, the input is a color image, convert to grayscale image segmentation, the output of grayscale images. The use of gray as the characteristics of each pixel clustering, due to light and other reasons, and sometimes should belong to an object pixel, its gray value will also be very different, may lead to clustering of the pixel error has occurred. in the segmentation results, the surface, there would be different from other pixel noise points, so , the algorithm Finally, the results of a median filter to eliminate noise, to the role of smoothing the image Platform: |
Size: 335872 |
Author:caoliang |
Hits:
Description: In this project ,we propose a color based segmentation method that uses the c means clustering technique to track tumor objects in magnetic resonance (MR) brain images. The key concept in this color based segmentation algorithm with k means means to convert a given gray level MR image in to a color space image and then separate the position of tumor objects from other items of an MR image by using c means clustering
And histogram clustering .Experiments demonstrates that the method can successfully achieve segmentation for MR brain images to help pathologists distinguish exactly lesion size and region.
-In this project ,we propose a color based segmentation method that uses the c means clustering technique to track tumor objects in magnetic resonance (MR) brain images. The key concept in this color based segmentation algorithm with k means means to convert a given gray level MR image in to a color space image and then separate the position of tumor objects from other items of an MR image by using c means clustering
And histogram clustering .Experiments demonstrates that the method can successfully achieve segmentation for MR brain images to help pathologists distinguish exactly lesion size and region.
Platform: |
Size: 2048 |
Author:pramod |
Hits:
Description: Kernel Entropy Component Analysis,KECA方法的作者R. Jenssen自己写的MATLAB代码，文章发表在2010年5月的IEEE TPAMI上面-Kernel Entropy Component Analysis, by R. Jenssen, published in IEEE TPAMI 2010.
We introduce kernel entropy component analysis (kernel ECA) as a new method for data transformation and dimensionality reduction. Kernel ECA reveals structure relating to the Renyi entropy of the input space data set, estimated via a kernel matrix using Parzen windowing. This is achieved by projections onto a subset of entropy preserving kernel principal component analysis (kernel PCA) axes. This subset does not need, in general, to correspond to the top eigenvalues of the kernel matrix, in contrast to the dimensionality reduction using kernel PCA. We show that kernel ECA may produce strikingly different transformed data sets compared to kernel PCA, with a distinct angle-based structure. A new spectral clustering algorithm utilizing this structure is developed with positive results. Furthermore, kernel ECA is shown to be an useful alternative for pattern denoising. Platform: |
Size: 3072 |
Author:johhnny |
Hits:
Description: 针对合成孔径雷达(SAR) 图像含有大量斑点噪声的特点,基于Contourlet 的多尺度、局部化、方向性和各向
异性等优点,并结合隐马尔科夫树( HMT) 模型和隐马尔科夫场(MRF) ,提出了一种基于Contourlet 域持续性和聚
集性的SAR 图像模糊融合分割算法。该算法有效捕获了Contourlet 子带的持续性和聚集性,并分别用HMT 和
MRF 来刻画,再依据模糊测度,将多尺度HMT 和MRF 有机融合,建立Contourlet 域HMT2MRF 融合模型,并导
出新模型下的最大后验概率(MAP) 分割公式。对实测SAR 图像进行了仿真,仿真结果和分析表明:与小波域上的
HMT2MRF 融合分割及Contourlet 域上HMT 和MRF 分割算法相比,该算法在抑制斑点噪声的同时,有效地提高
了SAR 图像的分割精度- In view of the speckle noise in the synthetic aperture radar (SAR) images , and based on the Contourlet′s
advantages of multiscale , localization , directionality , and anisot ropy , a new SAR image fusion segmentation
algorithm based on the pe rsis tence and clustering in the Contourlet domain is p roposed. The algorithm captures the
pe rsis tence and clus tering of the Contourlet t ransform , which is modeled by hidden Markov t ree (HMT) and Markov
random field (MRF) , respectively. Then , these two models are fused by fuzzy logic , resulting in a Contourlet
domain HMT2MRF fusion model . Finally , the maximum a poste rior (MAP) segmentation equation for the new fusion
model is deduced. The algorithm is used to emulate the real SAR images . Simulation results and analysis indicate that
the p roposed algorithm effectively reduces the influence of multiplicative speckle noise , imp roves the segmentation
accuracy and p rovides a bet te r visual quality for SAR images ove r the Platform: |
Size: 897024 |
Author:周二牛 |
Hits:
Description: Content-based medical image retrieval is now getting more and more attention in the
world, a feasible and efficient retrieving algorithm for clinical endoscopic images is urgently
required. Methods: Based on the study of single feature image retrieving techniques, including color
clustering, color texture and shape, a new retrieving method with multi-features fusion and relevance
feedback is proposed to retrieve the desired endoscopic images. Results: A prototype system is set
up to evaluate the proposed method’s performance and some evaluating parameters such as the
retrieval precision & recall, statistical average position of top 5 most similar image on various features, etc.
are therefore given. Conclusions: The algorithm with multi-features fusion and relevance feedback
gets more accurate and quicker retrieving capability than the one with single feature image retrieving
technique due to its flexible feature combination and interactive relevance feedback. Platform: |
Size: 359424 |
Author:gokul/goks |
Hits:
Description: 在CIEL*a*b*空间，
根据前面得出的图像分割的个数利用模糊聚类(FCM)对图像进行聚类，将得到的聚
类中心与从标准色卡中提出的绿色对应的L*、a*、b*的参考值进行比较，与其最接
近的类即为要提取的绿色作物目标
-Using fuzzy clustering (FCM) in CIEL* a* b* space, the number of the previously derived image segmentation image clustering, the cluster center will be raised from the standard color card corresponding to green L*, a*, b* of the reference values are compared, their closest classes is to be extracted green crops target Platform: |
Size: 6521856 |
Author:韩子晨 |
Hits:
Description: 针对光照变化对人脸检测及人眼定位的影响
，
提出一种基于肤色模型的人脸检测与人
眼定位方法
．
先对图像进行预处理
，
减少图像中的噪声
；
再将
ＲＧＢ
颜色空间转化到具有良好
肤色聚类特性的
ＹＣｂＣｒ
色彩空间
，
利用
Ｇａｕｓｓ
模型进行肤色建模
；
最后检测出人脸区域并确
定人眼位置
．
-The influence of illumination changes on human face detection and eye location
,
A face detection based on skin color model is proposed.
Eye location method
.
First image preprocessing
,
Reduce noise in images
Then
RGB
The color space is transformed to a good
Color clustering property
YCbCr
Color space
,
using
Gauss
Model for skin color modeling
Finally, the face region is detected.
Fixed eye position
. Platform: |
Size: 4096 |
Author:王文智 |
Hits:
Description: The influence of illumination changes on human face detection and eye location
,
A face detection based on skin color model is proposed.
Eye location method
.
First image preprocessing
,
Reduce noise in images
Then
RGB
The color space is transformed to a good
Color clustering property
YCbCr
Color space
,
using
Gauss
Model for skin color modeling
Finally, the face region is detected.
Fixed eye position
. Platform: |
Size: 2191360 |
Author:王文智 |
Hits:
Description: 此代码可以对图像很好的聚类，文件里面有原始图像，也有聚类后的图像，聚类的效果挺好的，大家可以看看(This code can make a good clustering of images. In the file, there are original images, and there are also images of clustering. The effect of clustering is good. You can have a look at it) Platform: |
Size: 118784 |
Author:宁采臣
|
Hits:
Search - clustering in images