专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US07805388B2 Method for feature selection in a support vector machine using feature ranking 失效
标题翻译：使用特征排序的支持向量机中特征选择的方法
公开(公告)号：US07805388B2
公开(公告)日：2010-09-28
申请号：US11928784
申请日：2007-10-30
申请人： Jason Weston , André Elisseeff , Bernhard Schölkopf , Fernando Perez-Cruz , Isabelle Guyon
发明人： Jason Weston , André Elisseeff , Bernhard Schölkopf , Fernando Perez-Cruz , Isabelle Guyon
IPC分类号： G06N7/00
CPC分类号： C12Q1/6883 , C12Q2600/112 , C12Q2600/158 , G06F19/20 , G06F19/24 , G06K9/6231 , G06K9/6269
摘要： In a pre-processing step prior to training a learning machine, pre-processing includes reducing the quantity of features to be processed using feature selection methods selected from the group consisting of recursive feature elimination (RFE), minimizing the number of non-zero parameters of the system (l0-norm minimization), evaluation of cost function to identify a subset of features that are compatible with constraints imposed by the learning set, unbalanced correlation score, transductive feature selection and single feature using margin-based ranking. The features remaining after feature selection are then used to train a learning machine for purposes of pattern classification, regression, clustering and/or novelty detection.
摘要翻译：在训练学习机之前的预处理步骤中，预处理包括使用从递归特征消除（RFE）中选出的特征选择方法来减少要处理的特征量的数量，使非零参数的数量最小化（10-norm minimization），评估成本函数以识别与由学习集施加的约束兼容的特征的子集，不平衡相关得分，转换特征选择和使用基于边缘的排名的单个特征。然后，特征选择之后剩余的特征用于训练学习机，用于模式分类，回归，聚类和/或新颖性检测。

2. 发明授权

US07475048B2 Pre-processed feature ranking for a support vector machine 失效
标题翻译：支持向量机的预处理功能排名
公开(公告)号：US07475048B2
公开(公告)日：2009-01-06
申请号：US10494876
申请日：2002-11-07
申请人： Jason Weston , André Elisseeff , Bernhard Schölkopf , Fernando Perez-Cruz , Isabelle Guyon
发明人： Jason Weston , André Elisseeff , Bernhard Schölkopf , Fernando Perez-Cruz , Isabelle Guyon
IPC分类号： G06F15/18
CPC分类号： G06K9/623 , G06K9/6269 , G06N99/005
摘要： A computer-implemented method is provided for ranking features within a large dataset containing a large number of features according to each feature's ability to separate data into classes. For each feature, a support vector machine separates the dataset into two classes and determines the margins between extremal points in the two classes. The margins for all of the features are compared and the features are ranked based upon the size of the margin, with the highest ranked features corresponding to the largest margins. A subset of features for classifying the dataset is selected from a group of the highest ranked features. In one embodiment, the method is used to identify the best genes for disease prediction and diagnosis using gene expression data from micro-arrays.
摘要翻译：提供了一种计算机实现的方法，用于根据每个特征将数据分离成类的能力，对包含大量特征的大型数据集中的特征进行排名。对于每个特征，支持向量机将数据集分为两类，并确定两类极值点之间的边距。比较所有功能的边距，并根据边距的大小对特征进行排名，排名最高的功能对应于最大的边距。从一组最高排名的特征中选择用于分类数据集的特征的子集。在一个实施方案中，该方法用于使用来自微阵列的基因表达数据鉴定用于疾病预测和诊断的最佳基因。

3. 发明申请

US20110106735A1 RECURSIVE FEATURE ELIMINATION METHOD USING SUPPORT VECTOR MACHINES 审中-公开
标题翻译：使用支持向量机的回归特征消除方法
公开(公告)号：US20110106735A1
公开(公告)日：2011-05-05
申请号：US12944197
申请日：2010-11-11
申请人： Jason Weston , André Elisseeff , Bernhard Schölkopf , Fernando Perez-Cruz , Isabelle Guyon
发明人： Jason Weston , André Elisseeff , Bernhard Schölkopf , Fernando Perez-Cruz , Isabelle Guyon
IPC分类号： G06F15/18
CPC分类号： G06K9/6231 , G06N20/00 , G06N20/10 , G16B25/00 , G16B40/00
摘要： Identification of a determinative subset of features from within a group of features is performed by training a support vector machine using training samples with class labels to determine a value of each feature, where features are removed based on their the value. One or more features having the smallest values are removed and an updated kernel matrix is generated using the remaining features. The process is repeated until a predetermined number of features remain which are capable of accurately separating the data into different classes. In some embodiments, features are eliminated by a ranking criterion based on a Lagrange multiplier corresponding to each training sample.
摘要翻译：通过使用具有类标签的训练样本来训练支持向量机来确定特征组中的特征的确定性子集来确定每个特征的值，其中基于其特征被去除。删除具有最小值的一个或多个特征，并且使用其余特征生成更新的内核矩阵。重复该过程，直到保持能够将数据精确地分离成不同类别的预定数量的特征。在一些实施例中，通过基于对应于每个训练样本的拉格朗日乘数的排序标准来消除特征。

4. 发明申请

US20110119213A1 SUPPORT VECTOR MACHINE - RECURSIVE FEATURE ELIMINATION (SVM-RFE) 有权
标题翻译：支持矢量机 - 恢复功能消除（SVM-RFE）
公开(公告)号：US20110119213A1
公开(公告)日：2011-05-19
申请号：US12957411
申请日：2010-12-01
申请人： André Elisseeff , Bernhard Schölkopf , Fernando Perez-Cruz
发明人： André Elisseeff , Bernhard Schölkopf , Fernando Perez-Cruz
IPC分类号： G06F15/18
CPC分类号： G06F19/24 , G06F19/20 , G06K9/6231
摘要： Identification of a determinative subset of features from within a group of features is performed by training a support vector machine using training samples with class labels to determine a value of each feature, where features are removed based on their the value. One or more features having the smallest values are removed and an updated kernel matrix is generated using the remaining features. The process is repeated until a predetermined number of features remain which are capable of accurately separating the data into different classes.
摘要翻译：通过使用具有类标签的训练样本来训练支持向量机来确定特征组中的特征的确定性子集来确定每个特征的值，其中基于其特征被去除。删除具有最小值的一个或多个特征，并且使用其余特征生成更新的内核矩阵。重复该过程，直到保持能够将数据精确地分离成不同类别的预定数量的特征。

5. 发明授权

US10032254B2 Method and device for recovering a digital image from a sequence of observed digital images 有权
公开(公告)号：US10032254B2
公开(公告)日：2018-07-24
申请号：US13825365
申请日：2011-09-28
申请人： Stefan Harmeling , Michael Hirsch , Suvrit Sra , Bernhard Schölkopf , Christian J. Schuler
发明人： Stefan Harmeling , Michael Hirsch , Suvrit Sra , Bernhard Schölkopf , Christian J. Schuler
IPC分类号： H04N5/232 , G06T5/00 , H04N5/357
摘要： A computer-implemented method for recovering a digital image (x) from a sequence of observed digital images (y1, . . . , yT), includes: obtaining an observed digital image (yt); estimating a point spread function (ft) based on the observed image (yt); estimating the recovered digital image (x), based on the estimated point spread function (ft) and the observed image (yt); and repeating the above steps. In order to correct optical aberrations of a lens, a point spread function of the lens may be used.

6. 发明授权

US08290568B2 Method for determining a property map of an object, particularly of a living being, based on at least a first image, particularly a magnetic resonance image 有权
标题翻译：基于至少第一图像，特别是磁共振图像来确定对象，特别是生物的属性图的方法
公开(公告)号：US08290568B2
公开(公告)日：2012-10-16
申请号：US12351429
申请日：2009-01-09
申请人： Bernd Pichler , Matthias Hofmann , Bernhard Schölkopf , Florian Steinke
发明人： Bernd Pichler , Matthias Hofmann , Bernhard Schölkopf , Florian Steinke
IPC分类号： A61B5/05
CPC分类号： G01R33/481 , A61B6/037 , G06T7/33 , G06T2207/10088 , G06T2207/30004 , G06T2207/30196
摘要： It is disclosed a system and method (12) for determining a property map (82) of an object, particularly a human being, based on at least a first image (84), particularly an magnetic resonance (MR) image, of the object. In the method (12), a structure of reference pairs is defined in a first step (96), wherein each reference pair (16-26) comprises at least two entries (62). The first entry represents a property value, particularly an attenuation value. The second entry (62) preferably represents a group of image points (67) belonging together, which is extracted particularly from MR images (28) and comprises an interesting image point corresponding to the property value. In another step (98) of the method (12) a plurality of training pairs (16-26) is provided. A structure of the training pairs (16-26) corresponds to the structure of reference pairs, and the entries of respective training pairs (16-26) are known. In another step (100) of the method (12), an assignment between the first entries and the other entries (62-66) of the training pairs (16-26) is determined by machine learning, thus allowing prediction of a property value (88) corresponding to an arbitrary point (90) of the first image (84).
摘要翻译：公开了一种用于基于至少第一图像（84），特别是磁共振（MR）图像来确定对象，特别是人类的属性图（82）的系统和方法（12）。在方法（12）中，在第一步骤（96）中定义参考对的结构，其中每个参考对（16-26）包括至少两个条目（62）。第一项表示属性值，特别是衰减值。第二条目（62）优选地表示属于一组的一组图像点（67），其特别是从MR图像（28）提取并包括与该属性值对应的有趣图像点。在方法（12）的另一步骤（98）中，提供了多个训练对（16-26）。训练对（16-26）的结构对应于参考对的结构，并且各训练对（16-26）的条目是已知的。在方法（12）的另一步骤（100）中，训练对（16-26）的第一条目和其他条目（62-66）之间的分配由机器学习确定，从而允许预测属性值（88）对应于第一图像（84）的任意点（90）。

7. 发明申请

US20130242129A1 METHOD AND DEVICE FOR RECOVERING A DIGITAL IMAGE FROM A SEQUENCE OF OBSERVED DIGITAL IMAGES 审中-公开
标题翻译：用于从观察数字图像的序列恢复数字图像的方法和装置
公开(公告)号：US20130242129A1
公开(公告)日：2013-09-19
申请号：US13825365
申请日：2011-09-28
申请人： Stefan Harmeling , Michael Hirsch , Suvrit Sra , Bernhard Schölkopf , Christian J. Schuler
发明人： Stefan Harmeling , Michael Hirsch , Suvrit Sra , Bernhard Schölkopf , Christian J. Schuler
IPC分类号： G06T5/00 , H04N5/232
CPC分类号： G06T5/001 , G06T5/003 , G06T2207/20201 , H04N5/23229 , H04N5/23267 , H04N5/3572
摘要： A computer-implemented method for recovering a digital image (x) from a sequence of observed digital images (y1, . . . , yT), includes: obtaining an observed digital image (yt); estimating a point spread function (ft) based on the observed image (yt); estimating the recovered digital image (x), based on the estimated point spread function (ft) and the observed image (yt); and repeating the above steps. In order to correct optical aberrations of a lens, a point spread function of the lens may be used.
摘要翻译：一种用于从观测数字图像序列（y1，...，yT）中恢复数字图像（x）的计算机实现方法包括：获得观察到的数字图像（yt）; 基于观察图像（yt）估计点扩散函数（ft）; 基于估计的点扩散函数（ft）和观察到的图像（yt）估计恢复的数字图像（x）; 并重复上述步骤。为了校正透镜的光学像差，可以使用透镜的点扩散函数。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式