公开(公告)号：US20100074551A1

公开(公告)日：2010-03-25

申请号：US12235488

申请日：2008-09-22

申请人： Zhouchen Lin ,  Wei Zhang

发明人： Zhouchen Lin ,  Wei Zhang

IPC分类号： G06K9/40

CPC分类号： G06K9/40 ,  G06T5/001 ,  G06T7/10 ,  G06T2207/20081

摘要： Partial differential equations (PDEs) are used in the invention for various problems in computer the vision space. The present invention provides a framework for learning a system of PDEs from real data to accomplish a specific vision task. In one embodiment, the system consists of two PDEs. One controls the evolution of the output. The other is for an indicator function that helps collect global information. Both PDEs are coupled equations between the output image and the indicator function, up to their second order partial derivatives. The way they are coupled is suggested by the shift and rotational invariance that the PDEs should hold. The coupling coefficients are learnt from real data via an optimal control technique. The invention provides learning-based PDEs that make a unified framework for handling different vision tasks, such as edge detection, denoising, segementation, and object detection.

摘要翻译： 局部微分方程（PDE）用于本发明的计算机视觉空间中的各种问题。 本发明提供了一种用于从实际数据学习PDE系统以完成特定视觉任务的框架。 在一个实施例中，系统由两个PDE组成。 一个控制输出的演变。 另一个是用于帮助收集全球信息的指标功能。 两个PDE是输出图像和指示符函数之间的耦合方程，直到它们的二阶偏导数。 它们耦合的方式是由PDE应该保持的移动和旋转不变性来提出的。 通过最优控制技术从实数数据中学习耦合系数。 本发明提供了基于学习的PDE，其构成用于处理不同视觉任务的统一框架，例如边缘检测，去噪，分割和对象检测。

公开(公告)号：US20100067799A1

公开(公告)日：2010-03-18

申请号：US12212222

申请日：2008-09-17

申请人： Guangcan Liu ,  Zhouchen Lin ,  Xiaoou Tang

发明人： Guangcan Liu ,  Zhouchen Lin ,  Xiaoou Tang

IPC分类号： G06K9/46

CPC分类号： G06K9/4647

摘要： A “globally invariant Radon feature transform,” or “GIRFT,” generates feature descriptors that are both globally affine invariant and illumination invariant. These feature descriptors effectively handle intra-class variations resulting from geometric transformations and illumination changes to provide robust texture classification. In general, GIRFT considers images globally to extract global features that are less sensitive to large variations of material in local regions. Geometric affine transformation invariance and illumination invariance is achieved by converting original pixel represented images into Radon-pixel images by using a Radon Transform. Canonical projection of the Radon-pixel image into a quotient space is then performed using Radon-pixel pairs to produce affine invariant feature descriptors. Illumination invariance of the resulting feature descriptors is then achieved by defining an illumination invariant distance metric on the feature space of each feature descriptor.

摘要翻译： “全局不变的氡特征变换”或“GIRFT”产生全局仿射不变和照明不变的特征描述符。 这些特征描述符有效地处理由几何变换和照明变化产生的类内变化，以提供鲁棒的纹理分类。 一般来说，GIRFT在全球范围内考虑图像，以提取对本地区域的大量材料较不敏感的全局特征。 通过使用Radon变换将原始像素表示的图像转换为氡像素图像来实现几何仿射变换不变性和照度不变性。 然后使用氡 - 像素对执行氡像素图像到商空间的规范投影，以产生仿射不变特征描述符。 然后通过在每个特征描述符的特征空间上定义照明不变距离度量来实现所得特征描述符的照明不变性。

公开(公告)号：US20090001165A1

公开(公告)日：2009-01-01

申请号：US11772069

申请日：2007-06-29

申请人： Chunhui Zhang ,  Zhouchen Lin ,  Zhengyou Zhang ,  Shi Han

发明人： Chunhui Zhang ,  Zhouchen Lin ,  Zhengyou Zhang ,  Shi Han

IPC分类号： G06K7/10

CPC分类号： G06K7/1093 ,  G06T3/608 ,  G06T2207/20061

摘要： Systems and methods for 2-D barcode recognition are described. In one aspect, the systems and methods use a charge coupled camera capturing device to capture a digital image of a 3-D scene. The systems and methods evaluate the digital image to localize and segment a 2-D barcode from the digital image of the 3-D scene. The 2-D barcode is rectified to remove non-uniform lighting and correct any perspective distortion. The rectified 2-D barcode is divided into multiple uniform cells to generate a 2-D matrix array of symbols. A barcode processing application evaluates the 2-D matrix array of symbols to present data to the user.

摘要翻译： 描述了用于二维条形码识别的系统和方法。 在一个方面，所述系统和方法使用电荷耦合的摄像机捕捉设备来捕获3-D场景的数字图像。 系统和方法评估数字图像，以从3-D场景的数字图像中定位和分割二维条形码。 二维条形码整流，以消除不均匀的照明并纠正任何透视失真。 经整流的二维条形码被分成多个均匀的单元格，以产生符号的二维矩阵阵列。 条形码处理应用程序评估符号的二维矩阵数组以向用户呈现数据。

公开(公告)号：US07439989B2

公开(公告)日：2008-10-21

申请号：US11276204

申请日：2006-02-17

申请人： Zhouchen Lin ,  Junfeng He ,  Xiaoou Tang

发明人： Zhouchen Lin ,  Junfeng He ,  Xiaoou Tang

IPC分类号： G06T15/00

CPC分类号： G06T1/0042 ,  G06T2201/0052 ,  H04N19/60 ,  H04N19/86

摘要： Systems and methods for detecting doctored JPEG images are described. In one aspect, a JPEG image is evaluated to determine if the JPEG image comprises double quantization effects of double quantized Discrete Cosine Transform coefficients. In response to results of these evaluation operations, the systems and methods determine whether the JPEG image has been doctored and identify any doctored portion.

摘要翻译： 描述用于检测编码的JPEG图像的系统和方法。 在一个方面，评估JPEG图像以确定JPEG图像是否包括双量化离散余弦变换系数的双量化效应。 响应于这些评估操作的结果，系统和方法确定JPEG图像是否被编辑并识别任何编辑部分。

公开(公告)号：US20080059867A1

公开(公告)日：2008-03-06

申请号：US11468609

申请日：2006-08-30

申请人： Zhouchen Lin ,  Qiang Wang ,  Chunhui Zhang

发明人： Zhouchen Lin ,  Qiang Wang ,  Chunhui Zhang

IPC分类号： H03M13/00

CPC分类号： H03M13/13

摘要： The present decoding technique provides an efficient technique for decoding linear block codes from multiple encoders. When an error in a code sequence is detected, the decoding technique estimates a confidence for each bit within the code sequence. Based on the confidence, a subset of bits within the code sequence is obtained. The subset of bits is then incrementally flipped to determine a set of modified code sequences. A syndrome is computed for each of the modified code sequences based on a preceding computed syndrome and an update vector.

摘要翻译： 本解码技术提供了用于从多个编码器解码线性块码的有效技术。 当检测到代码序列中的错误时，解码技术估计代码序列内每个位的置信度。 基于置信度，获得码序列内的比特的子集。 然后逐位翻转位的子集以确定一组修改的代码序列。 基于先前计算的综合征和更新向量，为每个修改的码序列计算校正子。

公开(公告)号：US20060262973A1

公开(公告)日：2006-11-23

申请号：US11132865

申请日：2005-05-19

申请人： Zhouchen Lin ,  Rongrong Wang ,  Xiaoou Tang ,  Heung-Yeung Shum

发明人： Zhouchen Lin ,  Rongrong Wang ,  Xiaoou Tang ,  Heung-Yeung Shum

IPC分类号： G06K9/62 ,  G06K9/00 ,  G06K15/00

CPC分类号： G06K9/00 ,  G06K9/00899

摘要： Embodiments of the invention determine whether an image has been altered. Sets of patches are selected in the image, and corresponding inverse response functions are provided to a support vector machine (SVM). The support vector machine is trained with exemplary normal and abnormal inverse response functions. Once trained, the support vector machine analyzes inverse response functions corresponding to a suspected image. The support vector machine determines if the inverse response functions are normal or abnormal by analyzing a set of features. In one embodiment, features include measures for monotonic characteristics, fluctuation characteristics, and divergence characteristics of the red, green, and blue components of a tuple. Each tuple of inverse response functions is associated with a set of patches selected in the image.

公开(公告)号：US08406556B2

公开(公告)日：2013-03-26

申请号：US12797859

申请日：2010-06-10

申请人： Jiaping Wang ,  Yue Dong ,  Xin Tong ,  Zhouchen Lin ,  Baining Guo

发明人： Jiaping Wang ,  Yue Dong ,  Xin Tong ,  Zhouchen Lin ,  Baining Guo

IPC分类号： G06K9/40

CPC分类号： G06T15/50

摘要： A “Scene Re-Lighter” provides various techniques for using an automatically reconstructed light transport matrix derived from a sparse sampling of images to provide various combinations of complex light transport effects in images, including caustics, complex occlusions, inter-reflections, subsurface scattering, etc. More specifically, the Scene Re-Lighter reconstructs the light transport matrix from a relatively small number of acquired images using a “Kernel Nyström” based technique adapted for low rank matrices constructed from sparsely sampled images. A “light transport kernel” is incorporated into the Nyström method to exploit nonlinear coherence in the light transport matrix. Further, an adaptive process is used to efficiently capture the sparsely sampled images from a scene. The Scene Re-Lighter is capable of achieving good reconstruction of the light transport matrix with only few hundred images to produce high quality relighting results. Further, the Scene Re-Lighter is also effective for modeling scenes with complex lighting effects and occlusions.

摘要翻译： 场景再打火机提供了使用从图像稀疏采样得到的自动重建光传输矩阵的各种技术，以提供图像中复杂光传输效应的各种组合，包括焦散，复杂遮挡，相互反射，地下散射等。 更具体地说，场景重新点亮器使用适用于由稀疏采样图像构成的低秩矩阵的基于内核Nyström的技术，从相对较少数量的获取图像重构光传输矩阵。 光传输核被并入Nyström方法，以利用光传输矩阵中的非线性相干性。 此外，使用自适应处理来有效地从场景捕获稀疏采样的图像。 场景重新打火机能够通过仅仅几百张图像实现光传输矩阵的良好重建，从而产生高品质的重视效果。 此外，场景重新打火机对于对具有复杂的照明效果和遮挡的场景进行建模也是有效的。

公开(公告)号：US08064697B2

公开(公告)日：2011-11-22

申请号：US11871764

申请日：2007-10-12

申请人： Deli Zhao ,  Zhouchen Lin ,  Xiaoou Tang

发明人： Deli Zhao ,  Zhouchen Lin ,  Xiaoou Tang

IPC分类号： G06K9/00 ,  G06T7/00

CPC分类号： G06K9/6248

摘要： Systems and methods perform Laplacian Principal Components Analysis (LPCA). In one implementation, an exemplary system receives multidimensional data and reduces dimensionality of the data by locally optimizing a scatter of each local sample of the data. The optimization includes summing weighted distances between low dimensional representations of the data and a mean. The weights of the distances can be determined by a coding length of each local data sample. The system can globally align the locally optimized weighted scatters of the local samples and provide a global projection matrix. The LPCA improves performance of such applications as face recognition and manifold learning.

摘要翻译： 系统和方法执行拉普拉斯主成分分析（LPCA）。 在一个实现中，示例性系统通过局部优化数据的每个局部采样的散射来接收多维数据并且降低数据的维度。 优化包括对数据的低维表示和平均值之间的加权距离求和。 距离的权重可以通过每个本地数据样本的编码长度来确定。 该系统可以对局部采样的局部优化加权散射进行全局对齐，并提供全局投影矩阵。 LPCA可以改善诸如面部识别和歧管学习等应用的性能。

公开(公告)号：US20100198902A1

公开(公告)日：2010-08-05

申请号：US12364533

申请日：2009-02-03

申请人： Xu Yang ,  Zhouchen Lin ,  Sijun Liu ,  Tianjun Ye

发明人： Xu Yang ,  Zhouchen Lin ,  Sijun Liu ,  Tianjun Ye

IPC分类号： G06F7/552

CPC分类号： G06F17/10

摘要： Described is a technology, such as implemented in a computational software program, by which a minimal polynomial is efficiently determined for a radical expression based upon its structure of the radical expression. An annihilation polynomial is found based upon levels of the radical to obtain roots of the radical. A numerical method performs a zero test or multiple zero tests to find the minimal polynomial. In one implementation, the set of roots corresponding to a radical expression is found. The annihilation polynomial is computed by grouping roots of the set according to their conjugation relationship and multiplying factor polynomials level by level. A selection mechanism selects the minimal polynomial based upon the annihilation polynomial's factors.

摘要翻译： 描述了一种技术，例如在计算软件程序中实现的技术，通过该技术，基于其基本表达式的结构，有效地确定基本表达式的最小多项式。 基于获得根的根的自由基的水平找到湮灭多项式。 数值方法执行零测试或多零测试以找到最小多项式。 在一个实现中，找到与激进表达相对应的一组根。 湮灭多项式通过根据它们的共轭关系和乘法因子多项式级别逐级分组的根来计算。 选择机制根据湮灭多项式的因素选择最小多项式。

公开(公告)号：US20100076723A1

公开(公告)日：2010-03-25

申请号：US12235927

申请日：2008-09-23

申请人： Wei Zhang ,  Zhouchen Lin ,  Xiaoou Tang

发明人： Wei Zhang ,  Zhouchen Lin ,  Xiaoou Tang

IPC分类号： G06F17/18 ,  G06F15/00 ,  G06F7/32 ,  G06F17/30

CPC分类号： G06F17/30598 ,  G06K9/6234

摘要： Tensor linear Laplacian discrimination for feature extraction is disclosed. One embodiment comprises generating a contextual distance based sample weight and class weight, calculating a within-class scatter using the at least one sample weight and a between-class scatter for multiple classes of data samples in a sample set using the class weight, performing a mode-k matrix unfolding on scatters and generating at least one orthogonal projection matrix.

摘要翻译： 公开了用于特征提取的张量线性拉普拉斯判别。 一个实施例包括生成基于上下文距离的样本权重和类权重，使用所述至少一个样本权重来计算类内散度，以及使用类权重在样本集合中的多类数据样本之间进行类间散射，执行 mode-k矩阵在散射上展开并生成至少一个正交投影矩阵。