搜索资源列表
-
1下载:
图像的特征用到了Dense Sift,通过Bag of Words词袋模型进行描述,当然一般来说是用训练集的来构建词典,因为我们还没有测试集呢。虽然测试集是你拿来测试的,但是实际应用中谁知道测试的图片是啥,所以构建BoW词典我这里也只用训练集。
其实BoW的思想很简单,虽然很多人也问过我,但是只要理解了如何构建词典以及如何将图像映射到词典维上去就行了,面试中也经常问到我这个问题,不知道你们都怎么用生动形象的语言来描述这个问题?
用BoW描述完图像之后,指的是将训练集以及测试集
-
-
0下载:
LTI库是一个面向对象的计算机视觉库写在C++视窗/余-VC++和Linux的/gcc.它提供了很多功能来解决数学问题、图像处理算法很多,有些工具和分类得多-LTI-Lib is an object oriented computer vision library written in C++ for Windows/MS-VC++ and Linux/gcc. It provides lots of functionality to solve mathematical problems, m
-
-
0下载:
基于kmeans的图像分割算法,首先得到图像,再确定分类,并计算各分类的均值,从而得到各个像素的分类。-Kmeans-based image segmentation algorithm, first of all get the image, and then determine the classification, and calculate the various categories mean, and thus obtain the classification of each pi
-
-
0下载:
对于图像中的不同水果,分别从颜色,形状等方面进行识别,并且分类。-For the image of different fruits, respectively from the color, shape, etc for identification and classification.
-
-
0下载:
图像识别算法。对图像中不同的对象进行分类识别。如三角开,四边形,五边形等。-Image recognition algorithm. The image objects of different classification. Such as the triangle open, quadrilateral, pentagon and so on.
-
-
4下载:
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-sca
-
-
0下载:
使用复杂网络提取图像边缘特征并进行识别的源代码,采用PCA_LDA算法对特征进行降维分类识别,识别效率很高。鲁棒性好-Extracted using image edge characteristics of complex networks and to identify the source code, using PCA_LDA algorithm to reduce the dimensions feature classification, identification with hi
-
-
0下载:
基于K均值算法和互信息熵差的算法,可以有效地确定分类数,从而该算法对医学图像进行自动优化分割。-K means algorithm based on entropy and mutual information algorithms, can effectively determine the classification number, so that the algorithm for automatic optimization of medical image segmentation.
-
-
0下载:
这是一个关于遥感图像非监督分类法的算法,希望对大家有所帮助-This is a remote sensing image on the unsupervised classification algorithm, we want to help
-
-
0下载:
Canny算子分类,实现对图像进行分割。代码进行过优化,有较高的实时性。-Canny operator classification, to achieve image segmentation. Code has been optimized, a higher real-time.
-
-
0下载:
对彩色图像进行分类识别,非常好用,识别率95%-Classification of color image recognition, is very easy to use, 95 recognition rate
-
-
0下载:
提供遥感图像的非监督分类,专业性比较强,可以提供给GIS RS GPS等地理专业的同志参看-Provide remote sensing image supervised classification of non-professional are relatively strong, could be made available to the GIS RS GPS and other geo-professional comrades see
-
-
0下载:
该软件需用Delphi7设计,采用灰度共生矩阵的方法对肝脏超声图像进行纹理特征提取。通过神经网络进行分类处理。-The software required Delphi7 design, the use of gray-scale co-occurrence matrix method of liver ultrasound image texture feature extraction. Through the neural network classification.
-
-
0下载:
Adaboost算法的基本思想是:利用大量的分类能力一般的弱分类器(weaker
ifier)通过一定的方法叠加(boost)起来,构成一个分类能力很强的强分类器
眼eClassifier),再将若干个强分类器串联成为分级分类器(ClassifierCaseade)
图像搜索检测。本文就是利用Adaboost算法将由类haar特征生成的弱分类器
成为强分类器,再将强分类器串联成为分级分类器。
-Adaboost algorithm basic idea is: the abi
-
-
0下载:
图象模式识别的两个程序:分类识别与聚类识别,代码编写条理,简洁,思路清晰,易懂。
一定可以运行。-Image pattern recognition of two procedures: Classification and Clustering identification, coding coherent, concise, clear, easy to understand. Will be able to run.
-
-
0下载:
用于图像识别的核主分量识别子程序,不包括分类器程序-For image recognition to identify the nuclear principal component subprogram does not include classification program
-
-
0下载:
此软件的功能是对遥感图像进行非监督分类。实现算法是k均值-This software is a remote sensing image for non-supervised classification. K-means algorithm implementation are
-
-
1下载:
c#下实现图像颜色k均值聚类
将像素点颜色分类并合并-c# implementation of color image k-means clustering to classification of color pixels and merge
-
-
0下载:
-
-
0下载:
预处理时中值滤波然后二值化对图像A进行二值化得到二值化图像B -The next section concerns the principal component analysis on image plates and the classification efficiency obtained
-