公开(公告)号：US20130322698A1

公开(公告)日：2013-12-05

申请号：US13876823

申请日：2010-10-01

申请人： Anders Modén

发明人： Anders Modén

IPC分类号： G06T7/20

CPC分类号： G06T7/2033 ,  G06T7/246 ,  G06T7/285 ,  G06T7/77 ,  G06T2207/10012 ,  G06T2207/20076 ,  G06T2207/30244

摘要： A method and apparatus for determining position and orientation enabling navigation of an object using image data from at least a first and a second 2D image from at least one camera mounted on said object. The method comprises the steps of: correcting images from one or several cameras and from at least a first and a second 2D image for their respective radial distortion and other measurable effects which result in poor image precision; matching 2D image items in and between at least a first and second 2D image; calculating a fundamental matrix by using correlated image points from at least a first and a second 2D image; calculating and extracting estimated first rotation and translation values from the fundamental matrix using single value decomposition (SVD) based on information from at least a first and a second 2D image; iterating more accurate final rotation and translation values by using the LevenBerg-Marquard algorithm and determining the position and orientation of said object.

摘要翻译： 一种用于确定使用来自安装在所述对象上的至少一个相机的至少第一和第二2D图像的图像数据来导航对象的位置和方向的方法和装置。 该方法包括以下步骤：从一个或多个照相机和至少第一和第二2D图像中校正图像，以获得它们各自的径向失真和其它可测量的影响，这导致图像精度差; 在至少第一和第二2D图像之中和之间匹配2D图像项目; 通过使用来自至少第一和第二2D图像的相关图像点来计算基本矩阵; 基于来自至少第一和第二2D图像的信息，使用单值分解（SVD）从基本矩阵计算和提取估计的第一旋转和平移值; 通过使用LevenBerg-Marquard算法迭代更准确的最终旋转和平移值，并确定所述对象的位置和方向。

公开(公告)号：US20130272581A1

公开(公告)日：2013-10-17

申请号：US13877252

申请日：2010-10-01

申请人： Anders Modén

发明人： Anders Modén

IPC分类号： G06T7/00

CPC分类号： G06T7/0042 ,  G06T7/70 ,  G06T7/73 ,  G06T2207/30244

摘要： A method and apparatus for determining a position and attitude of at least one camera by calculating and extracting estimated rotation and translation values from an estimated fundamental matrix based on information from at least a first and second 2D image. Variable substitution is utilized to strengthen derivatives and provide a more rapid convergence. A solution is provided for solving position and orientation from correlated point features in images using a method that solves for both rotation and translation simultaneously.

摘要翻译： 一种用于通过基于来自至少第一和第二2D图像的信息的估计基本矩阵计算和提取估计旋转和平移值来确定至少一个照相机的位置和姿态的方法和装置。 利用变量替代来加强衍生品并提供更快速的收敛。 提供了一种解决方案，用于使用同时解决旋转和平移的方法来解决图像中相关点特征的位置和方向。

公开(公告)号：US09020204B2

公开(公告)日：2015-04-28

申请号：US13876823

申请日：2010-10-01

申请人： Anders Modén

发明人： Anders Modén

IPC分类号： G06K9/00 ,  G06T7/20 ,  G06T7/00

CPC分类号： G06T7/2033 ,  G06T7/246 ,  G06T7/285 ,  G06T7/77 ,  G06T2207/10012 ,  G06T2207/20076 ,  G06T2207/30244

摘要： A method and apparatus for determining position and orientation enabling navigation of an object using image data from at least a first and a second 2D image from at least one camera mounted on said object. The method comprises the steps of: correcting images from one or several cameras and from at least a first and a second 2D image for their respective radial distortion and other measurable effects which result in poor image precision; matching 2D image items in and between at least a first and second 2D image; calculating a fundamental matrix by using correlated image points from at least a first and a second 2D image; calculating and extracting estimated first rotation and translation values from the fundamental matrix using single value decomposition (SVD) based on information from at least a first and a second 2D image; iterating more accurate final rotation and translation values by using the LevenBerg-Marquard algorithm and determining the position and orientation of said object.

摘要翻译： 一种用于确定使用来自安装在所述对象上的至少一个相机的至少第一和第二2D图像的图像数据来导航对象的位置和方向的方法和装置。 该方法包括以下步骤：从一个或多个照相机和至少第一和第二2D图像中校正图像，以获得它们各自的径向失真和其它可测量的影响，这导致图像精度差; 在至少第一和第二2D图像之中和之间匹配2D图像项目; 通过使用来自至少第一和第二2D图像的相关图像点来计算基本矩阵; 基于来自至少第一和第二2D图像的信息，使用单值分解（SVD）从基本矩阵计算和提取估计的第一旋转和平移值; 通过使用LevenBerg-Marquard算法迭代更准确的最终旋转和平移值，并确定所述对象的位置和方向。

公开(公告)号：US08953847B2

公开(公告)日：2015-02-10

申请号：US13877252

申请日：2010-10-01

申请人： Anders Modén

发明人： Anders Modén

IPC分类号： G06K9/00 ,  G06T7/00

CPC分类号： G06T7/0042 ,  G06T7/70 ,  G06T7/73 ,  G06T2207/30244

摘要： A method and apparatus for determining a position and attitude of at least one camera by calculating and extracting estimated rotation and translation values from an estimated fundamental matrix based on information from at least a first and second 2D image. Variable substitution is utilized to strengthen derivatives and provide a more rapid convergence. A solution is provided for solving position and orientation from correlated point features in images using a method that solves for both rotation and translation simultaneously.

摘要翻译： 一种用于通过基于来自至少第一和第二2D图像的信息的估计基本矩阵计算和提取估计旋转和平移值来确定至少一个照相机的位置和姿态的方法和装置。 利用变量替代来加强衍生品并提供更快速的收敛。 提供了一种解决方案，用于使用同时解决旋转和平移的方法来解决图像中相关点特征的位置和方向。

公开(公告)号：US20130208009A1

公开(公告)日：2013-08-15

申请号：US13876822

申请日：2010-10-01

申请人： Anders Modén

发明人： Anders Modén

IPC分类号： G06T5/00

CPC分类号： G06T5/006 ,  G06T7/593 ,  G06T2207/20076

摘要： A method and apparatus for generating and optimizing a fundamental matrix for a first 2D image and a second 2D image to obtain the relative geometrical information between said two 2D images for points in the two 2D images that correspond to a mutual 3D point. According to the method, the geometrical projection errors in the correspondence points are used to select correct and accurate inliers. This method and apparatus provides a more accurate and precise fundamental matrix than conventional methods.

摘要翻译： 一种用于生成和优化用于第一2D图像和第二2D图像的基本矩阵的方法和装置，以获得对应于相互3D点的两个2D图像中的点的所述两个2D图像之间的相对几何信息。 根据该方法，对应点中的几何投影误差用于选择正确和准确的内容。 该方法和装置提供比传统方法更准确和精确的基本矩阵。

公开(公告)号：US07813543B2

公开(公告)日：2010-10-12

申请号：US11570130

申请日：2005-06-08

申请人： Anders Modén

发明人： Anders Modén

IPC分类号： G06T17/00

CPC分类号： G06T7/593 ,  G06T7/30

摘要： The present invention relates to automatic modeling of a physical scene. At least two images (I1, I2) of the scene are received, which are taken from different angles and/or positions. A matching module (130) matches image objects in the first image (I1) against image objects in the second image (I2), by first loading pixel values for at least one first portion of the first image (I1) into an artificial neural network (133). Then, the artificial neural network (133) scans the second image (I2) in search of pixels representing a respective second portion corresponding to each of the at least one first portion; determines a position of the respective second portion upon fulfillment of a match criterion; and produces a representative matching result (M12). Based on the matching result (M12), a first calculation module (140) calculates a fundamental matrix (F12), which defines a relationship between the first and second images (I1, I2). Based on the fundamental matrix (F12), in turn, a second calculation module (150) calculates a depth map (D12), which describes distance differences between a set of image points in the first image (I1) and a corresponding set of image points in the second image (I2). Finally, the depth map (D12) constitutes a basis for a synthetic model of the scene.

摘要翻译： 本发明涉及物理场景的自动建模。 接收场景的至少两个图像（I1，I2），其从不同的角度和/或位置取得。 匹配模块（130）通过首先将第一图像（I1）的至少一个第一部分的像素值加载到人造神经网络中来匹配第一图像（I1）中的图像对象与第二图像（I2）中的图像对象 （133）。 然后，人造神经网络（133）扫描第二图像（I2），以搜索表示对应于至少一个第一部分中的每一个的相应第二部分的像素; 在实现匹配标准时确定相应的第二部分的位置; 并产生代表性匹配结果（M12）。 基于匹配结果（M12），第一计算模块（140）计算定义第一和第二图像（I1，I2）之间关系的基本矩阵（F12）。 基于基本矩阵（F12），第二计算模块（150）依次计算描述第一图像（I1）中的一组图像点与相应图像集合之间的距离差的深度图（D12） 点在第二个图像（I2）。 最后，深度图（D12）构成了场景合成模型的基础。