公开(公告)号：US11645829B2

公开(公告)日：2023-05-09

申请号：US16815397

申请日：2020-03-11

申请人： VY Corporation

发明人： Thomas Martel

IPC分类号： G06V10/44 ,  G06T7/13 ,  G06T7/90 ,  G06T7/12 ,  G06T7/181 ,  G06V10/46 ,  G06T7/136

CPC分类号： G06V10/457 ,  G06T7/12 ,  G06T7/13 ,  G06T7/136 ,  G06T7/181 ,  G06T7/90 ,  G06V10/471

摘要： A Shape Based Modeling Segmentation fits generated Bézier curves on to edges parsed from an object in an image, identifies the Bézier curves in predefined proximity having at least one of a geometric relationship and a reporting relationship with other Bézier curves in the predefined proximity; generates MetaBézier curves from the identified Bézier curves; and connects the MetaBézier curves to identify the object in the image.

公开(公告)号：US20200294259A1

公开(公告)日：2020-09-17

申请号：US16815397

申请日：2020-03-11

申请人： VY Corporation

发明人： Thomas Martel

IPC分类号： G06T7/564 ,  G06T7/13 ,  G06T7/12 ,  G06T7/529 ,  G06T7/90

摘要： A Shape Based Modeling Segmentation fits generated Bézier curves on to edges parsed from an object in an image, identifies the Bézier curves in predefined proximity having at least one of a geometric relationship and a reporting relationship with other Bézier curves in the predefined proximity; generates MetaBézier curves from the identified Bézier curves; and connects the MetaBézier curves to identify the object in the image.

公开(公告)号：US07742082B1

公开(公告)日：2010-06-22

申请号：US11131449

申请日：2005-05-17

申请人： Thomas Martel

发明人： Thomas Martel

IPC分类号： H04N5/235

CPC分类号： H04N5/35581 ,  H04N5/235 ,  H04N5/2355

摘要： A system and method of increasing the sensitivity of an optical detector formed from a two-dimensional array of sensor elements, each of said sensor elements having a dynamic range. A gradient-detection algorithm is applied to an image received from the optical detector in order to identify a shape and position of a first subset of pixels in the image. The first subset of pixels corresponds to an object represented in the image. A determination is made whether one or more of the sensor elements corresponding in shape and position to the first subset of pixels have been exposed outside of their dynamic range. If one or more sensor elements have been overexposed, at least one of the gain or offset associated with the one or more of the sensor elements is adjusted, such that the one or more sensor elements are exposed within their dynamic range. The adjustment of the gain or offset of the one or more sensor elements is performed independently of any gain or offset adjustment applied to other sensor elements in the optical detector.

摘要翻译： 一种增加由传感器元件的二维阵列形成的光学检测器的灵敏度的系统和方法，每个所述传感器元件具有动态范围。 梯度检测算法被应用于从光学检测器接收的图像，以便识别图像中的第一像素子集的形状和位置。 第一像素子集对应于图像中表示的对象。 确定在像素的第一子集的形状和位置对应的传感器元件中的一个或多个是否已经暴露在它们的动态范围之外。 如果一个或多个传感器元件已经曝光过度，则调整与一个或多个传感器元件相关联的增益或偏移中的至少一个，使得一个或多个传感器元件在其动态范围内被暴露。 一个或多个传感器元件的增益或偏移的调节独立于应用于光学检测器中的其它传感器元件的任何增益或偏移调整。

公开(公告)号：US08942486B1

公开(公告)日：2015-01-27

申请号：US13550829

申请日：2012-07-17

申请人： Thomas Martel ,  John T. Freyhof

发明人： Thomas Martel ,  John T. Freyhof

IPC分类号： G06K9/64 ,  G06K9/62

CPC分类号： G06K9/468 ,  G06K9/48 ,  G06T3/4053 ,  G06T7/593

摘要： Elements of an electronic image are organized into groups to obtain descriptive data associated with the electronic image. A wide field view of the electronic image is obtained from a first component and a higher resolution image of a selected portion of the wide field view of the electronic image is obtained from a second component to resolve ambiguity associated the selected portion of the wide field view of the electronic image. At least one primitive is formed using pixels of the electronic image, where the primitive is a curve primitive or a region primitive. The at least one primitive is analyzed using at least one level in a ladder of abstraction to organize elements of the electronic image into groups from which descriptive data can be obtained about at least one of objects or activities associated with the electronic image.

摘要翻译： 将电子图像的元素组织成组以获得与电子图像相关联的描述性数据。 从第一组件获得电子图像的宽视场图，并且从第二组件获得电子图像的宽视野视图的选定部分的较高分辨率图像，以解决与广视场视图的所选部分相关联的模糊度 的电子图像。 使用电子图像的像素形成至少一个原语，其中原语是曲线原语或区域原语。 使用抽象梯形图中的至少一个级别来分析至少一个原语，以将电子图像的元素组合成可以从哪个关于与电子图像相关联的对象或活动中的至少一个获得描述性数据的组。

公开(公告)号：US08326786B1

公开(公告)日：2012-12-04

申请号：US12487476

申请日：2009-06-18

申请人： Thomas Martel ,  John T. Freyhof

发明人： Thomas Martel ,  John T. Freyhof

IPC分类号： G06F15/00

CPC分类号： G06K9/48 ,  G06K2009/485

摘要： An apparatus and method is disclosed for acquiring an electronic image and forming at least one Grenze Set including pixels of the electronic image. A decision tree is used to apply vocabulary and rules associated with a primitive to evaluate pixels of the Grenze Set. The pixels of the Grenze Set are explained by re-building the Grenze Set using a set of sub-primitives. Higher order analysis are applied to the Grenze Set according to a ladder of abstraction to assemble pixels into at least one of objects or activities that are meaningful to a user.

摘要翻译： 公开了一种用于获取电子图像并形成包括电子图像的像素的至少一个Grenze集的装置和方法。 决策树用于应用与原语相关联的词汇和规则，以评估Grenze集合的像素。 通过使用一组子图元重新构建Grenze集来解释Grenze集合的像素。 根据抽象梯子将高阶分析应用于Grenze集合，以将像素组装成对用户有意义的对象或活动中的至少一个。

公开(公告)号：US07734112B1

公开(公告)日：2010-06-08

申请号：US11637634

申请日：2006-12-12

申请人： Thomas Martel

发明人： Thomas Martel

IPC分类号： G06K9/40

CPC分类号： G06K9/48

摘要： A system and method for identifying an object in an electronically acquired image that includes at least one two-dimensional array of pixels. A plurality of one-dimensional gradient runs oriented along a common direction in the two-dimensional array of pixels is identified. A second-order set of gradient runs is formed by selecting a group of previously identified one-dimensional gradient runs. Each of the one-dimensional gradient runs in the group has a pixel that is offset along an axis perpendicular to the common direction from a pixel in a neighboring one-dimensional gradient run in the group. The object is identified in the array using the second-order set of gradient runs.

摘要翻译： 一种用于识别包括至少一个二维像素阵列的电子获取图像中的对象的系统和方法。 识别在二维像素阵列中沿着公共方向定向的多个一维梯度运行。 通过选择一组先前识别的一维梯度运行来形成二阶梯度运行。 组中的每个一维梯度运行具有沿着在组中运行的相邻一维梯度中的像素沿着垂直于公共方向的轴偏移的像素。 该对象使用二阶梯度运算集在数组中标识。

公开(公告)号：US20230419569A1

公开(公告)日：2023-12-28

申请号：US18213331

申请日：2023-06-23

申请人： VY Corporation

发明人： Thomas Martel ,  John Freyhof

IPC分类号： G06T11/20 ,  G06T7/13 ,  G06V10/44

CPC分类号： G06T11/203 ,  G06V2201/07 ,  G06V10/457 ,  G06T7/13

摘要： A Shape-Based Modeling Segmentation (SBMS) system is provided for differentiating compact objects having finite boundaries from extended objects. The SBMS system includes a Bézier curve system to fit generated Bézier curves on edges parsed from an object in an image. The SBMS system also includes a CPELinear platform to identify the Bézier curves in a predefined proximity, the Bézier curves having at least one of a geometric relationship and a reporting relationship with other Bézier curves in the predefined proximity, detect linear objects having slightly curved or linear boundaries identified from the Bézier curves, and connect the Bézier curves to identify the object in the image.

公开(公告)号：US09544565B1

公开(公告)日：2017-01-10

申请号：US14269391

申请日：2014-05-05

申请人： Thomas Martel

发明人： Thomas Martel

IPC分类号： H04N13/00

CPC分类号： H04N13/106 ,  G01B11/25 ,  G01B21/04 ,  H04N13/20

摘要： Data for making calculations from a three dimensional observation is derived from a recording device. The data is combined with information developed by a three dimensional remote sensing platform to create measurement points in space for an object. Descriptive information from at least one object model is used to direct at least one of a resolution resource, results gained from group measurements and an object-deployed resolution asset. Order is thereafter found in two dimensional to three dimensional observations in a subject area.

摘要翻译： 从记录装置导出用于从三维观察进行计算的数据。 该数据与由三维遥感平台开发的信息相结合，以创建物体的空间测量点。 使用来自至少一个对象模型的描述性信息来指导分辨率资源，从组测量获得的结果和对象部署的分辨率资产中的至少一个。 此后，在主题区域中的二维至三维观察中发现顺序。

公开(公告)号：US08718320B1

公开(公告)日：2014-05-06

申请号：US12012863

申请日：2008-02-05

申请人： Thomas Martel ,  John Freyhof

发明人： Thomas Martel ,  John Freyhof

IPC分类号： G06K9/00

CPC分类号： G06K9/4604 ,  G06K9/4638 ,  G06K9/6202 ,  G06K2209/25

摘要： Vision pattern recognition is used to discover the presence, position, orientation, and movement of physical objects within electronic representations of data such as images based on the objects' appearance, according to certain embodiments.

摘要翻译： 根据某些实施例，使用视觉模式识别来发现物理对象在诸如基于对象外观的图像的数据的电子表示内的存在，位置，取向和移动。