登录
首页 » matlab » hyper_SAE

hyper_SAE

于 2015-08-15 发布 文件大小:10548KB
0 125
下载积分: 1 下载次数: 171

代码说明:

  用SAE 深度学习方法做的高光谱图像分析的一个例子。里面有sae方法还有高光谱图像的特征提取过程。(One example of the depth of learning to do with SAE hyperspectral image analysis. There are sae method has hyperspectral image feature extraction process.)

文件列表:

SAE
...\checkNumericalGradient.m,2026,2014-04-15
...\computeNumericalGradient.m,1314,2014-04-15
...\display_network.m,2945,2014-04-15
...\IMAGES.mat,20971720,2011-01-03
...\initializeParameters.m,635,2014-04-15
...\minFunc
...\.......\ArmijoBacktrack.m,3251,2011-01-03
...\.......\autoGrad.m,807,2011-01-03
...\.......\autoHess.m,901,2011-01-03
...\.......\autoHv.m,317,2011-01-03
...\.......\autoTensor.m,870,2011-01-03
...\.......\callOutput.m,385,2011-01-03
...\.......\conjGrad.m,1845,2011-01-03
...\.......\dampedUpdate.m,995,2011-01-03
...\.......\example_minFunc.m,2421,2011-01-03
...\.......\example_minFunc_LR.m,1604,2011-01-03
...\.......\isLegal.m,107,2011-01-03
...\.......\lbfgs.m,924,2011-01-03
...\.......\lbfgsC.c,2408,2011-01-03
...\.......\lbfgsC.mexa64,7707,2011-01-03
...\.......\lbfgsC.mexglx,7733,2011-01-03
...\.......\lbfgsC.mexmac,9500,2011-01-03
...\.......\lbfgsC.mexmaci,12660,2011-01-03
...\.......\lbfgsC.mexmaci64,8800,2011-01-03
...\.......\lbfgsC.mexw32,7168,2011-01-03
...\.......\lbfgsC.mexw64,9728,2011-01-03
...\.......\lbfgsUpdate.m,614,2011-01-03
...\.......\logistic
...\.......\........\LogisticDiagPrecond.m,417,2011-01-03
...\.......\........\LogisticHv.m,216,2011-01-03
...\.......\........\LogisticLoss.m,659,2011-01-03
...\.......\........\mexutil.c,1154,2011-01-03
...\.......\........\mexutil.h,317,2011-01-03
...\.......\........\mylogsumexp.m,227,2011-01-03
...\.......\........\repmatC.c,3965,2011-01-03
...\.......\........\repmatC.dll,7680,2011-01-03
...\.......\........\repmatC.mexglx,20682,2011-01-03
...\.......\........\repmatC.mexmac,10000,2011-01-03
...\.......\mchol.m,1287,2011-01-03
...\.......\mcholC.c,4190,2011-01-03
...\.......\mcholC.mexmaci64,13184,2011-01-03
...\.......\mcholC.mexw32,8192,2011-01-03
...\.......\mcholC.mexw64,12288,2011-01-03
...\.......\mcholinc.m,564,2011-01-03
...\.......\minFunc.m,43634,2011-01-03
...\.......\minFunc_processInputOptions.m,3704,2011-01-03
...\.......\polyinterp.m,4217,2011-01-03
...\.......\precondDiag.m,42,2011-01-03
...\.......\precondTriu.m,51,2011-01-03
...\.......\precondTriuDiag.m,60,2011-01-03
...\.......\rosenbrock.m,1074,2011-01-03
...\.......\taylorModel.m,677,2011-01-03
...\.......\WolfeLineSearch.m,11478,2011-01-03
...\sampleIMAGES.m,2512,2014-04-15
...\sparseAutoencoderCost.m,5822,2014-04-15
...\train.m,6156,2014-04-17

下载说明:请别用迅雷下载,失败请重下,重下不扣分!

发表评论

0 个回复

  • fourcomponontdecomposition
    极化SAR图像Four component 目标分解结果,使用数据为Flevoland地区部分数据,可用于极化SAR图像分类(Four component decomposition result of PolSAR image, the used dataset is the subarea of Flevoland region, it can be used as the classification features for PolSAR image classification.)
    2018-05-25 16:14:45下载
    积分:1
  • EDLines 快速直线检测算子
    EDLines是一种快速直线检测算子,2012年在ICCV上提出,也是目前处理直线检测最快的算法之一,该算法包含三个步骤:(1)边缘提取:利用Edge Drawing (ED)算法[28,29]从灰度图像中提取边缘片段;(2)线段检测:利用最小二乘法提取直线段;(3)线段确认:遵循Helmholtz定律,从已提取的直线段中摒弃虚假线段。EDLines算法的优越性得益于Edge Drawing (ED)算法能够从灰度图像中准确、快速、稳定地提取出光滑、完整的边缘片段。Edge Drawing (ED)算法包含以下步骤: (1)采用size=5*5,σ=1的高斯核对灰度图像进行平滑滤波,去除噪声; (2)采用一种常用的梯度算子,如Prewitt、Sobel或Scharr等计算平滑后图像中每个像素点的梯度幅度和梯度方向; (3)将梯度图中邻域内幅度最大的像素点标记为锚点,这些锚点成为图像边缘点(edge elements)的概率很大; (4)将相邻的锚点连接成边缘线。从一个锚点起始,ED利用相邻像素的梯度幅度和方向在梯度为最大值的锚点之间游走。 (Edge Drawing (ED) is our recently-proposed, novel, fast edge detection algorithm. What makes ED stand out the existing edge detectors, e.g., Canny, is the following: While the other edge detectors give out a binary edge image as output, where the etected edge pixels are usually independent, discontinuous entities ED produces a set of edge segments, which are clean, contiguous, i.e., connected, chains of edge pixels. Thus, while the output of the other edge detectors requires urther processing to generate potential object boundaries, which may not even be possible or result in inaccuracies ED not only produces perfectly connected object boundaries by default, but it also achieves this in blazing speed compared to other edge detector.)
    2021-03-22 16:29:16下载
    积分:1
  • Tarel algorithm
    经典的Tarel去雾算法,可用于图像去雾的指标性能对比。(It is the classical dehazing algorithm, and it is can be used comparison with other algorithms.)
    2020-06-26 16:00:01下载
    积分:1
  • 压缩感知中利用增广拉格朗日方程解最小稀疏正则化的恢复算法 DAL
    压缩感知中利用增广拉格朗日方程解最小稀疏正则化的恢复算法(DAL solves the dual problem of (1) via the augmented Lagrangian method (see Bertsekas 82). It uses the analytic expression (and its derivatives) of the following soft-thresholding operation, which can be computed for L1 and grouped L1 (and many other) sparsity inducing regularizers. If you are interested in our algorithm please find more details in our technical report or in my talk at Optimization for Machine Learning Workshop (NIPS 2009).)
    2011-04-25 10:19:38下载
    积分:1
  • EM-MRF-ICM
    基于EM-马尔科夫随机场-ICM的图像分割程序,很好的学习代码(EM-based Markov random-ICM image segmentation procedures, good learning code)
    2013-11-14 23:10:39下载
    积分:1
  • RGBImage
    利用双线性插值和最近邻插值实现RGB图像的缩放,利用MATLAB编写(The use of bilinear interpolation and nearest neighbor interpolation to achieve RGB image scaling, using MATLAB to prepare)
    2009-10-31 19:29:03下载
    积分:1
  • Edge_Enhancement_Adaptive_Anisotropic_Diffusion
    说明:  完成基于各向异性的自适应超声图像去噪和边缘增强方法,参考Adaptive Anisotropic Diffusion for Ultrasonic Image Denoising and Edge Enhancement.pdf,加入了m参数(The completion of ultrasound-based adaptive anisotropic image denoising and edge enhancement method, reference Adaptive Anisotropic Diffusion for Ultrasonic Image Denoising and Edge Enhancement.pdf, joined the m parameters)
    2009-08-18 09:42:32下载
    积分:1
  • DCT
    该函数用来实现快速离散余弦变换。该函数利用2N点的快速付立叶变换来实现离散余弦变换(This function is used to implement fast discrete cosine transform. 2N points of the function using the fast Fourier transform to achieve the discrete cosine transform)
    2010-11-15 09:53:27下载
    积分:1
  • ICP-point-cloud-registration
    三维激光点云配准是点云三维建模的关键问题之一。经典的 ICP 算法对点云初始位置要求较高且配准 效率较低,提出了一种改进的 ICP 点云配准算法。该算法首先利用主成分分析法实现点云的初始配准,获得较好 的点云初始位置,然后在经典 ICP 算法的基础上,采用 k - d tree 结构实现加速搜索,并利用方向向量夹角阈值去除 错误点对,提高算法的效率。实验表明,本算法流程在保证配准精度的前提下,显著提高了配准效率。 (Three-dimensional laser point cloud registration is one of the key three-dimensional point cloud model. High classical ICP algorithm to the initial position of the point cloud registration requirements and low efficiency, proposed an improved ICP point cloud registration algorithm. Firstly, the use of principal component analysis of the initial point cloud registration, get a better initial position of the point cloud, then the basis of classical ICP algorithm using k- d tree structure to achieve speed up the search, and using the direction vector angle the removal of the threshold point error and improve the efficiency of the algorithm. Experiments show that the algorithm processes to ensure the accuracy of registration under the premise, significantly improve the efficiency of registration.)
    2016-08-01 10:34:57下载
    积分:1
  • 几何校正
    实现对辐射校正后的科学数据几何校正以及对角度数据校正,输出控制点文件(The calibration of the scientific data after radiation correction, the correction of the angle data and the output control point files are realized)
    2017-09-26 16:55:41下载
    积分:1
  • 696522资源总数
  • 104049会员总数
  • 30今日下载