How to extract optimal composite attributes from a variety of conventional seismic attributes to detect reservoir features is a reservoir predication key,which is usually solved by reducing dimensionality.Principle co...How to extract optimal composite attributes from a variety of conventional seismic attributes to detect reservoir features is a reservoir predication key,which is usually solved by reducing dimensionality.Principle component analysis（PCA） is the most widely-used linear dimensionality reduction method at present.However,the relationships between seismic attributes and reservoir features are non-linear,so seismic attribute dimensionality reduction based on linear transforms can＇t solve non-linear problems well,reducing reservoir prediction precision.As a new non-linear learning method,manifold learning supplies a new method for seismic attribute analysis.It can discover the intrinsic features and rules hidden in the data by computing low-dimensional,neighborhood-preserving embeddings of high-dimensional inputs.In this paper,we try to extract seismic attributes using locally linear embedding（LLE）,realizing inter-horizon attributes dimensionality reduction of 3D seismic data first and discuss the optimization of its key parameters.Combining model analysis and case studies,we compare the dimensionality reduction and clustering effects of LLE and PCA,both of which indicate that LLE can retain the intrinsic structure of the inputs.The composite attributes and clustering results based on LLE better characterize the distribution of sedimentary facies,reservoir,and even reservoir fluids.展开更多
A fault detection method based on incremental locally linear embedding(LLE)is presented to improve fault detecting accuracy for satellites with telemetry data.Since conventional LLE algorithm cannot handle incremental...A fault detection method based on incremental locally linear embedding(LLE)is presented to improve fault detecting accuracy for satellites with telemetry data.Since conventional LLE algorithm cannot handle incremental learning,an incremental LLE method is proposed to acquire low-dimensional feature embedded in high-dimensional space.Then,telemetry data of Satellite TX-I are analyzed.Therefore,fault detection are performed by analyzing feature information extracted from the telemetry data with the statistical indexes T2 and squared prediction error(SPE)and SPE.Simulation results verify the fault detection scheme.展开更多
现有全局优化算法都使用不同范数约束输出图像梯度来实现图像平滑,但会牺牲图像中的弱结构信息来达到较好的平滑性能,导致输出图像出现颜色失真和细节模糊的情况。针对上述问题,提出一种基于LLE的边缘保持图像平滑算法(edge preserving ...现有全局优化算法都使用不同范数约束输出图像梯度来实现图像平滑,但会牺牲图像中的弱结构信息来达到较好的平滑性能,导致输出图像出现颜色失真和细节模糊的情况。针对上述问题,提出一种基于LLE的边缘保持图像平滑算法(edge preserving image smoothing algorithm based on LLE,Ep-LLE),引入局部线性嵌入(LLE)的思想作为优化函数的正则化项并采用L_(2)范数进行惩罚。该方法利用图像局部区域内像素存在的相互关系,通过约束局部相似以实现图像平滑任务。最后通过各个算法的实验对比验证,基于LLE的边缘保持图像平滑算法能在实现图像边缘保持平滑的同时,保留图像局部结构特征,并有效避免区域内颜色一致导致的边缘阶梯状现象,避免图像颜色失真。展开更多
LLE(Locally Linear Embedding)算法是一种较好的流形学习算法,但它只能以批处理的方式进行.只要有新的样本加入,就必须重作该算法的全部内容,而原处理结果被全部丢弃.本文提出了一种基于正交迭代的增量LLE算法,能有效地利用前面的处理...LLE(Locally Linear Embedding)算法是一种较好的流形学习算法,但它只能以批处理的方式进行.只要有新的样本加入,就必须重作该算法的全部内容,而原处理结果被全部丢弃.本文提出了一种基于正交迭代的增量LLE算法,能有效地利用前面的处理结果,实现增量处理.实验表明该算法是有效的.展开更多
在人像识别方面,传统的特征提取方法大都是线性的,不能很好地保持样本的拓扑结构。支持向量机能提高学习的泛化能力,防止过学习,是一种很好的分类器。为此,提出一种增强的LLE(Locally Linear Em- bedding)和SVM(support Vector Machine...在人像识别方面,传统的特征提取方法大都是线性的,不能很好地保持样本的拓扑结构。支持向量机能提高学习的泛化能力,防止过学习,是一种很好的分类器。为此,提出一种增强的LLE(Locally Linear Em- bedding)和SVM(support Vector Machine)结合的人像识别方法,采用PCA(Principal Component Analysis)与LLE相结合算法,对光照归一化处理过的人脸图像进行特征提取,利用SVM的分类机制对人脸图像样本集进行训练和识别。在ORL(Olivetti Research Laboratory)人脸数据库上实验表明,该算法稳健、快速,识别率达到了90%以上。展开更多
结合人脸考勤系统项目实例介绍了用VFW(Video for Windows)实现视频图像采集、编辑和LLE(Locally Linear Enbed-ding)对采集的图像进行特征提取技术,并利用该技术建立了SCU_TS人脸库。结果表明,该技术实用、可靠,为视频应用程序的开发...结合人脸考勤系统项目实例介绍了用VFW(Video for Windows)实现视频图像采集、编辑和LLE(Locally Linear Enbed-ding)对采集的图像进行特征提取技术,并利用该技术建立了SCU_TS人脸库。结果表明,该技术实用、可靠,为视频应用程序的开发提供了一种行之有效的方法。展开更多
针对人脸检测问题的特点,提出一种基于改进型深度LLE(Locally Linear Embedding)算法和随机森林相结合的人脸检测算法。首先,通过采集图像的深度信息,结合图像的颜色信息,构建三维图像信息数据库,再通过改进的LLE算法得到最优降维结果,...针对人脸检测问题的特点,提出一种基于改进型深度LLE(Locally Linear Embedding)算法和随机森林相结合的人脸检测算法。首先,通过采集图像的深度信息,结合图像的颜色信息,构建三维图像信息数据库,再通过改进的LLE算法得到最优降维结果,按一定比例选取训练集,输入随机森林算法建立数据分类器;最后,将测试集输入到训练完成的分类器中,实现人脸图像的检测。选取Yale,JAFFE 2类数据集与传统算法进行对比实验,验证算法的优越性和可行性。实验结果表明:所提出的算法可以有效地完成人脸检测,检测率高于传统算法7%左右。展开更多
基金National Key Science & Technology Special Projects(Grant No.2008ZX05000-004)CNPC Projects(Grant No.2008E-0610-10).
文摘How to extract optimal composite attributes from a variety of conventional seismic attributes to detect reservoir features is a reservoir predication key,which is usually solved by reducing dimensionality.Principle component analysis（PCA） is the most widely-used linear dimensionality reduction method at present.However,the relationships between seismic attributes and reservoir features are non-linear,so seismic attribute dimensionality reduction based on linear transforms can＇t solve non-linear problems well,reducing reservoir prediction precision.As a new non-linear learning method,manifold learning supplies a new method for seismic attribute analysis.It can discover the intrinsic features and rules hidden in the data by computing low-dimensional,neighborhood-preserving embeddings of high-dimensional inputs.In this paper,we try to extract seismic attributes using locally linear embedding（LLE）,realizing inter-horizon attributes dimensionality reduction of 3D seismic data first and discuss the optimization of its key parameters.Combining model analysis and case studies,we compare the dimensionality reduction and clustering effects of LLE and PCA,both of which indicate that LLE can retain the intrinsic structure of the inputs.The composite attributes and clustering results based on LLE better characterize the distribution of sedimentary facies,reservoir,and even reservoir fluids.
基金supported by the Fundamental Research Funds for the Central Universities(No.2016083)
文摘A fault detection method based on incremental locally linear embedding(LLE)is presented to improve fault detecting accuracy for satellites with telemetry data.Since conventional LLE algorithm cannot handle incremental learning,an incremental LLE method is proposed to acquire low-dimensional feature embedded in high-dimensional space.Then,telemetry data of Satellite TX-I are analyzed.Therefore,fault detection are performed by analyzing feature information extracted from the telemetry data with the statistical indexes T2 and squared prediction error(SPE)and SPE.Simulation results verify the fault detection scheme.
文摘现有全局优化算法都使用不同范数约束输出图像梯度来实现图像平滑,但会牺牲图像中的弱结构信息来达到较好的平滑性能,导致输出图像出现颜色失真和细节模糊的情况。针对上述问题,提出一种基于LLE的边缘保持图像平滑算法(edge preserving image smoothing algorithm based on LLE,Ep-LLE),引入局部线性嵌入(LLE)的思想作为优化函数的正则化项并采用L_(2)范数进行惩罚。该方法利用图像局部区域内像素存在的相互关系,通过约束局部相似以实现图像平滑任务。最后通过各个算法的实验对比验证,基于LLE的边缘保持图像平滑算法能在实现图像边缘保持平滑的同时,保留图像局部结构特征,并有效避免区域内颜色一致导致的边缘阶梯状现象,避免图像颜色失真。
文摘LLE(Locally Linear Embedding)算法是一种较好的流形学习算法,但它只能以批处理的方式进行.只要有新的样本加入,就必须重作该算法的全部内容,而原处理结果被全部丢弃.本文提出了一种基于正交迭代的增量LLE算法,能有效地利用前面的处理结果,实现增量处理.实验表明该算法是有效的.
文摘在人像识别方面,传统的特征提取方法大都是线性的,不能很好地保持样本的拓扑结构。支持向量机能提高学习的泛化能力,防止过学习,是一种很好的分类器。为此,提出一种增强的LLE(Locally Linear Em- bedding)和SVM(support Vector Machine)结合的人像识别方法,采用PCA(Principal Component Analysis)与LLE相结合算法,对光照归一化处理过的人脸图像进行特征提取,利用SVM的分类机制对人脸图像样本集进行训练和识别。在ORL(Olivetti Research Laboratory)人脸数据库上实验表明,该算法稳健、快速,识别率达到了90%以上。
文摘结合人脸考勤系统项目实例介绍了用VFW(Video for Windows)实现视频图像采集、编辑和LLE(Locally Linear Enbed-ding)对采集的图像进行特征提取技术,并利用该技术建立了SCU_TS人脸库。结果表明,该技术实用、可靠,为视频应用程序的开发提供了一种行之有效的方法。
文摘针对人脸检测问题的特点,提出一种基于改进型深度LLE(Locally Linear Embedding)算法和随机森林相结合的人脸检测算法。首先,通过采集图像的深度信息,结合图像的颜色信息,构建三维图像信息数据库,再通过改进的LLE算法得到最优降维结果,按一定比例选取训练集,输入随机森林算法建立数据分类器;最后,将测试集输入到训练完成的分类器中,实现人脸图像的检测。选取Yale,JAFFE 2类数据集与传统算法进行对比实验,验证算法的优越性和可行性。实验结果表明:所提出的算法可以有效地完成人脸检测,检测率高于传统算法7%左右。
