公开(公告)号：US07092554B2

公开(公告)日：2006-08-15

申请号：US09846718

申请日：2001-05-01

Applicant: Shoupu Chen ,  Lawrence A. Ray

Inventor： Shoupu Chen ,  Lawrence A. Ray

IPC: G06K9/00 ,  G06K9/62

CPC classification number: G06K9/00597 ,  G06K9/00268 ,  G06T7/74

Abstract: A digital image processing method for locating eyes and mouth in a digital face image. The method includes the steps of detecting iris colored pixels in the digital face image; grouping the iris colored pixels into clusters; detecting eye positions using the iris colored pixels; identifying salient pixels relating to a facial feature in the digital face image; generating a signature curve using the salient pixels; and using the signature curve and the eye positions to locate a mouth position. In a preferred embodiment, a summation of squared difference method is used to detect the eye positions. In another preferred embodiment, the eyes and mouth positions are validated using statistics.

公开(公告)号：US06925195B2

公开(公告)日：2005-08-02

申请号：US10156530

申请日：2002-05-28

Applicant: Nathan D. Cahill ,  Lawrence A. Ray ,  Joseph F. Revelli, Jr.

Inventor： Nathan D. Cahill ,  Lawrence A. Ray ,  Joseph F. Revelli, Jr.

IPC: G01S7/36 ,  G01S7/497 ,  G01S17/36 ,  G01S17/89 ,  G06T7/20 ,  H04N5/232 ,  G06K9/00

CPC classification number: G06T7/20 ,  G01S7/497 ,  G01S17/36 ,  G01S17/89 ,  H04N5/23248 ,  H04N5/23264

Abstract: A method for generating a stabilized three-dimensional image from a scannerless range imaging system comprises the steps of acquiring a bundle of three or more phase offset images corresponding to modulated illumination reflected from a scene, whereby one or more of the phase offset images includes image motion relative to another phase offset image; searching for a stable chain of phase offset images in the image bundle, wherein a stable chain is a collection of images that is obtained by testing the phase offset images against a confidence measure that separates less severe correctable image motion from more severe image motion and then populating the stable chain only with those phase offset images that meet the confidence measure for correctable image motion; correcting for the image motion in the stable chain of images if a stable chain of at least three phase offset images is found to exist; and computing a stabilized three-dimensional image using the phase offset images from the stable chain of phase offset images.

Abstract translation: 一种用于从无扫描仪范围成像系统生成稳定三维图像的方法包括以下步骤：获取与从场景反射的调制照明相对应的三个或更多个相位偏移图像的束，由此相位偏移图像中的一个或多个包括图像 相对于另一相位偏移图像的运动; 在图像束中搜索稳定的相位偏移图像链，其中稳定链是通过相对于将较不严格的可校正图像运动与更严重的图像运动分离的可信度测量来测量相位偏移图像获得的图像的集合，然后 仅使用满足可校正图像运动的置信度的相位偏移图像来填充稳定链; 如果存在至少三个相位偏移图像的稳定链，则校正图像的稳定链中的图像运动; 以及使用来自稳定的相位偏移图像链的相位偏移图像来计算稳定的三维图像。

公开(公告)号：US06920237B2

公开(公告)日：2005-07-19

申请号：US09740422

申请日：2000-12-19

Applicant: Shoupu Chen ,  Lawrence A. Ray

Inventor： Shoupu Chen ,  Lawrence A. Ray

IPC: G06T1/00 ,  G06K9/00 ,  G06T7/00 ,  H04N1/46 ,  H04N1/60

CPC classification number: G06K9/00604 ,  G06K9/00228

Abstract: A digital image processing method is taught for detecting human irises in a digital image. The method comprises the steps measuring the red intensity of the pixels in the image, determining the probability that each pixel is an iris based upon the red intensity of the pixel, determining the probability that each pixel is not an iris based upon the red intensity of the pixel; and determining whether the pixel is an iris by analyzing the relationship between the probability that the pixel is an iris and the probability that the pixel is not an iris. In one embodiment of the present invention, the determination as to whether a pixel is an iris pixel is then made based upon the application of a Bayes model to the probability that the pixel is not an iris, the probability of the occurrence of an iris in the identified region and probability of the occurrence of a non-iris pixel in the identified region. In another embodiment of the present invention, the method comprises the steps of finding an oval shaped skin color region, detecting iris color pixels in the oval shaped skin color region, detecting iris color pixels in the oval shaped region using a Bayes model and locating eye positions based upon the detected iris color pixels.A computer program product for performing these methods is also taught.

Abstract translation: 教授数字图像处理方法来检测数字图像中的人体虹膜。 该方法包括测量图像中的像素的红色强度的步骤，基于像素的红色强度确定每个像素是虹膜的概率，根据红色强度确定每个像素不是虹膜的概率 像素; 以及通过分析像素是虹膜的概率与像素不是虹膜的概率之间的关系来确定像素是否为虹膜。 在本发明的一个实施例中，关于像素是否是虹膜像素的确定是基于贝叶斯模型应用于像素不是虹膜的可能性，发生虹膜的概率 所识别的区域和在所识别的区域中出现非虹膜像素的概率。 在本发明的另一实施例中，该方法包括以下步骤：发现椭圆形皮肤颜色区域，检测椭圆形皮肤颜色区域中的虹膜彩色像素，使用贝叶斯模型和定位眼睛检测椭圆形区域中的虹膜彩色像素 基于检测到的虹膜颜色像素的位置。 还教导了用于执行这些方法的计算机程序产品。

公开(公告)号：US06795090B2

公开(公告)日：2004-09-21

申请号：US10010896

申请日：2001-11-13

Applicant: Nathan D. Cahill ,  Lawrence A. Ray

Inventor： Nathan D. Cahill ,  Lawrence A. Ray

IPC: G09G536

CPC classification number: G06T15/205

Abstract: Intermediate panoramic images are each generated from two original panoramic images, where each panoramic image provides a 360 degree field of view of a scene from different nodal points. First, pairs of planar images are derived, where each pair corresponds to original planar views derived from respective portions of the original panoramic images, where a similar area of the scene is visible from each pair of planar views. Then, an intermediate planar image is generated from each pair of planar images, and the resulting series of intermediate planar images are cylindrically concatenated to form the intermediate panoramic image.

Abstract translation: 中间全景图像分别由两个原始全景图像生成，其中每个全景图像提供来自不同节点的场景的360度视场。 首先，导出成对的平面图像，其中每对对应于从原始全景图像的相应部分导出的原始平面视图，其中场景的类似区域从每对平面视图可见。 然后，从每对平面图像生成中间平面图像，并且所得到的一系列中间平面图像被圆柱形连接以形成中间全景图像。

公开(公告)号：US06792134B2

公开(公告)日：2004-09-14

申请号：US09740562

申请日：2000-12-19

Applicant: Shoupu Chen ,  Lawrence A. Ray

Inventor： Shoupu Chen ,  Lawrence A. Ray

IPC: G06K900

CPC classification number: G06K9/00604 ,  G06T7/74 ,  G06T7/90 ,  G06T2207/30216

Abstract: The present invention comprises a digital image processing method for detecting human eyes in a digital image. This method comprises the steps of: detecting iris pixels in the image; clustering the iris pixels, and selecting at least one of the following methods to identify eye positions: applying geometric reasoning to detect eye positions using the iris pixel clusters; applying a summation of squared difference method using the iris pixel clusters to detect eye positions; and applying a summation of squared difference method to detect eye positions from the pixels in the image. The method applied is selected on the basis of the number of iris pixel clusters. In another embodiment, the present invention also comprises a computer program product.

Abstract translation: 本发明包括一种用于在数字图像中检测人眼的数字图像处理方法。 该方法包括以下步骤：检测图像中的虹膜像素; 聚焦虹膜像素，并选择以下方法中的至少一种来识别眼睛位置：使用虹膜像素簇来应用几何推理来检测眼睛位置; 使用虹膜像素簇应用平方差法的求和来检测眼睛位置; 并且应用平方差方法的求和来从图像中的像素检测眼睛位置。 基于虹膜像素簇的数量选择应用的方法。 在另一个实施例中，本发明还包括计算机程序产品。

公开(公告)号：US06349174B1

公开(公告)日：2002-02-19

申请号：US09572522

申请日：2000-05-17

Applicant: Lawrence A. Ray ,  Louis R. Gabello

Inventor： Lawrence A. Ray ,  Louis R. Gabello

IPC: G03B1336

CPC classification number: H04N5/2226 ,  G01S17/36 ,  G01S17/89 ,  H04N13/254

Abstract: A color scannerless range imaging system includes an illumination system for controllably illuminating a scene with modulated illumination and an image responsive element for capturing image light from the scene, including the modulated image light. The system establishes a primary optical path for directly image light toward the image responsive element. A beamsplitter located in the primary optical path separates the image light into two channels, a first channel including an infrared component and a second channel including a color texture component, whereby one of the channels traverses a secondary optical path distinct from the primary path. A modulating element is operative in the first channel to receive the infrared component and a modulating signal, and to generate a processed infrared component with phase data indicative of range information. An optical network is provided in the secondary optical path for recombining the secondary optical path into the primary optical path such that the processed infrared component and the color texture component are directly to the image responsive element.

Abstract translation: 一种无彩色无扫描范围成像系统，包括用于以调制照明可控地照亮场景的照明系统和用于从包括调制图像光的场景捕获图像光的图像响应元件。 该系统建立用于直接将图像光朝向图像响应元件的主光路。 位于主光路中的分束器将图像光分离成两个通道，包括红外分量的第一通道和包括颜色纹理分量的第二通道，由此一个通道穿过与主路径不同的次级光路。 调制元件在第一通道中操作以接收红外分量和调制信号，并且产生具有指示范围信息的相位数据的经处理的红外分量。 在第二光路中提供光网络，用于将次光路重新组合到主光路中，使得经处理的红外分量和颜色纹理成分直接映射到图像响应元件。

公开(公告)号：US06288776B1

公开(公告)日：2001-09-11

申请号：US09449101

申请日：1999-11-24

Applicant: Nathan D. Cahill ,  Lawrence A. Ray

Inventor： Nathan D. Cahill ,  Lawrence A. Ray

IPC: G01C308

CPC classification number: G01S17/89 ,  G01B11/24 ,  G01S17/06

Abstract: A method of unambiguous range estimation is provided for use with a range imaging system that derives phase images from image pixels of a digital image. The method involves generating (a) a first phase image having one or more ambiguous phase intervals and (b) at least one additional phase image that is generated by shifting the phase intervals of the first phase image. Then at least one region of intersection between phase intervals in the two phase images is identified. Finally, the phase of at least one of the ambiguous phase intervals in the first phase image is adjusted based on values of the phase of the image pixels that belong to the region of intersection. As a result, the phase adjustment unwraps the phase ambiguity in the phase intervals of the first phase image.

Abstract translation: 提供了一种无歧义范围估计的方法，用于从数字图像的图像像素导出相位图像的范围成像系统。 该方法包括产生（a）具有一个或多个不明确的相位间隔的第一相位图像和（b）通过移动第一相位图像的相位间隔而产生的至少一个附加相位图像。 然后识别两相图像中的相位间隔之间的至少一个交叉区域。 最后，基于属于交叉区域的图像像素的相位值来调整第一相位图像中至少一个模糊相位间隔的相位。 结果，相位调整在第一相位图像的相位间隔中展开相位模糊度。

公开(公告)号：US5764231A

公开(公告)日：1998-06-09

申请号：US884001

申请日：1992-05-15

Applicant: Lawrence A. Ray ,  Richard N. Ellson

Inventor： Lawrence A. Ray ,  Richard N. Ellson

IPC: G03B35/00 ,  G06T15/20 ,  G06T17/00

CPC classification number: H04N1/00201 ,  G06T15/20

Abstract: A method and apparatus in which a geometric graphic image of an object to be rendered as a depth image is created. The viewpoint of the viewer of the depth image is determined by the user. Once the viewpoint is determined the number of views is automatically determined along with the spacing between or positions of the views. The system adjusts the aspect ratio of each view and rotates the object prior to rendering. The rendered views are then stored as electronic interleaved images which are used to produce a depth image, such as a lenticular print.

Abstract translation: 创建要呈现为深度图像的对象的几何图形图像的方法和装置。 深度图像的观看者的观点由用户确定。 一旦确定了视点，视图的数量自动地与视图的间隔或位置一起确定。 系统调整每个视图的纵横比，并在渲染之前旋转对象。 渲染的视图然后被存储为用于产生诸如透镜状印刷品之类的深度图像的电子交错图像。

公开(公告)号：US5673376A

公开(公告)日：1997-09-30

申请号：US885706

申请日：1992-05-19

Applicant: Lawrence A. Ray ,  Richard N. Ellson ,  David J. Kroth

Inventor： Lawrence A. Ray ,  Richard N. Ellson ,  David J. Kroth

IPC: G09G5/36 ,  G06T3/40 ,  G06T3/60 ,  G06T11/20 ,  G06T17/40 ,  H04N1/387 ,  G06T15/10

CPC classification number: G06T3/40

Abstract: A system and method that generates computer graphic images using anti-aliasing rendering to create sets of offset images each of a lower resolution than the resolution desired. The anti-aliasing rendering is performed using offset weights that effectively cause a graphic model to be rendered at different rendering points for each low resolution image which are offset between low resolution images. The lower resolution offset images are then interlaced or combined to produce the desired high resolution image.

Abstract translation: 一种使用抗混叠渲染来生成计算机图形图像的系统和方法，以创建比所需分辨率更低的分辨率的偏移图像集合。 使用偏移权重执行反混叠渲染，该偏移权重有效地使得在低分辨率图像之间偏移的每个低分辨率图像的不同渲染点处呈现图形模型。 然后将较低分辨率的偏移图像交织或组合以产生所需的高分辨率图像。

公开(公告)号：US5673320A

公开(公告)日：1997-09-30

申请号：US392713

申请日：1995-02-23

Applicant: Lawrence A. Ray ,  Richard N. Ellson

Inventor： Lawrence A. Ray ,  Richard N. Ellson

IPC: A61B5/117 ,  G06T7/00 ,  G07D7/00 ,  G07F7/12 ,  H04L9/00

CPC classification number: G07D7/00 ,  G07F7/08 ,  G07F7/122

Abstract: Multiple validations of printed documents incorporating image information and authorizing data on a printed document assist in the printed document validation process. This technique requires the authorized document holder to have an image identification accompany the application or production of the document. Image information is converted to a storable image that is used in one of a plurality of validating schemes that assures that the presenter of the printed document is not a substitute. Such schemes included visual comparison of the printed document presenter and extracted image information and validation that the data has not been altered. Non-reversible encryption of the data, as it is read from the document at the document presentation site is used to formulate encoded authorization data that is then compared against like encoded authorized document holder data stored at a centrally located data base.

Abstract translation: 包含图像信息和授权打印文档数据的打印文档的多个验证有助于打印文档验证过程。 这种技术要求授权的文档持有者在文档的应用或生产过程中具有图像识别。 图像信息被转换成可以确保打印文档的呈现者不是替代的多个验证方案之一中使用的可存储图像。 这样的方案包括打印文档呈现者的视觉比较和提取的图像信息以及数据未被改变的验证。 数据的不可逆加密（从文档呈现站点的文档中读取）被用于制定编码的授权数据，然后与存储在位于中心位置的数据库中的类似编码的授权文件夹数据进行比较。