公开(公告)号：US06249602B1

公开(公告)日：2001-06-19

申请号：US09086326

申请日：1998-05-28

Applicant: David J. Michael ,  Juha Koljonen ,  Paul Dutta-Choudhury

Inventor： David J. Michael ,  Juha Koljonen ,  Paul Dutta-Choudhury

IPC: G06K900

CPC classification number: G06K9/3216 ,  G06K9/6203 ,  G06T7/30 ,  G06T7/60

Abstract: An efficient and reliable method and apparatus is disclosed that finds a reference point of an object profile within an image when the object is of an unknown size. The object profile is modeled using a synthetic labeled-projection model, which in conjunction with the image, is projected over a portion of the image of the object profile to derive a histogram. The histogram is normalized and a maximum of a first derivative of the histogram is defined for that position. The position of the labeled-projection model is moved relative to the image, and the process is repeated until a selected portion of the image has been examined. The first derivative of the normalized labeled projection is greatest when a feature of the image and the feature denoted by a specific synthetic label of the labeled-projection model are aligned. The method and apparatus can locate the center of the object with reliability, because use of the labeled-projection model and the histogram minimizes the * effects of image artifacts. Further, the method decreases computational time, and thus, increases performance speed.

Abstract translation: 公开了一种有效且可靠的方法和装置，当物体具有未知尺寸时，找到图像内的对象轮廓的参考点。 使用合成的标注投影模型对对象轮廓进行建模，该模型与图像一起投影在对象轮廓的图像的一部分上以导出直方图。 对直方图进行归一化，并为该位置定义直方图的一阶导数的最大值。 标记投影模型的位置相对于图像移动，并重复该过程直到已经检查图像的所选部分。 当图像的特征和由标记投影模型的特定合成标签表示的特征对齐时，归一化标记投影的一阶导数最大。 该方法和装置可以可靠地定位对象的中心，因为使用标注投影模型和直方图可以最大限度地减小图像伪影的*效应。 此外，该方法减少了计算时间，从而提高了性能速度。

公开(公告)号：US20100046824A1

公开(公告)日：2010-02-25

申请号：US12611886

申请日：2009-11-03

Applicant: Paul Dutta-Choudhury

Inventor： Paul Dutta-Choudhury

IPC: G06K9/00

CPC classification number: G06K9/00134

Abstract: Methods, computer software, and systems for analyzing a biological specimen (e.g., a cytological specimen) on a slide are provided. Magnified image data frames of the biological specimen are taken at different locations on the slide. An object that is not entirely contained within at least one of the image data frames is identified, and complementary portions of the object respectively located in different ones of the image data frames are matched. A fully integrated object is created from the matched object portions. Attributes of the integrated object are then analyzed.

Abstract translation: 提供了用于分析幻灯片上的生物样本（例如，细胞学标本）的方法，计算机软件和系统。 在幻灯片上的不同位置拍摄生物样本的放大图像数据帧。 识别不完全包含在至少一个图像数据帧内的对象，并且分别位于不同图像数据帧中的对象的互补部分相匹配。 从匹配的对象部分创建完全集成的对象。 然后分析集成对象的属性。

公开(公告)号：US06697535B1

公开(公告)日：2004-02-24

申请号：US09447726

申请日：1999-11-24

Applicant: Paul Dutta-Choudhury

Inventor： Paul Dutta-Choudhury

IPC: G06K940

CPC classification number: G06K9/4633

Abstract: Methods and apparatuses are disclosed for refining groupings of edge points that represent contours in an image. The methods and apparatuses decrease data dispersion and data quantization effects. The methods and apparatuses are particularly useful for accurate and robust detection of straight line-segment features contained in noisy, cluttered imagery occurring in industrial machine vision applications. Additionally, a measurement criterion of the quality of the detected line segments is introduced.

Abstract translation: 公开了用于对表示图像中的轮廓的边缘点进行精细分组的方法和装置。 方法和装置降低数据色散和数据量化效应。 这些方法和装置对于在工业机器视觉应用中发生的嘈杂，杂乱的图像中包含的直线段特征的精确和鲁棒检测特别有用。 另外，引入检测线段的质量的测量标准。

公开(公告)号：US06289492B1

公开(公告)日：2001-09-11

申请号：US09303214

申请日：1999-04-30

Applicant: Paul Dutta-Choudhury ,  Len Picard ,  Yasunari Tosa

Inventor： Paul Dutta-Choudhury ,  Len Picard ,  Yasunari Tosa

IPC: G06F1750

CPC classification number: H01L21/67138 ,  G01N21/95684 ,  H01L21/681 ,  H01L2224/48091 ,  H01L2224/49171 ,  H01L2924/00014

Abstract: Methods and apparatuses are disclosed for defining a locus of viable points on an elongated object, such as a locus of viable bond points on a lead of a lead frame. The locus guides the bond point positioning of a wire bonder so that the number of unacceptable wire bonds is reduced. The locus is a region adapted to receive a bond, and is defined such that for any bond centered therein, the probability of forming an unacceptable bond is reduced by offsetting the locus from at least a portion of the lead edge(s) by a minimum-offset distance. The minimum-offset distance is based on various factor(s), such as user tolerance and bond width. Further, the definition of the locus is independent of lead shape, and consequently works with a variety of lead shapes. The locus can also have various shapes for a single lead, and the shapes can vary depending on the application. The method and apparatus herein disclosed describe a variety of methods to detect the leads and define the size, position, and shape of the locus during training. Preferably, the lead tip is also detected so that the locus can be defined based on an offset from the front of the lead. During operation, characteristics of the locus and/or the target bond point are used to relocate the lead and refine the target bond/point before bonding. Alternative embodiments disclosed include inspecting the leads.

Abstract translation: 公开了用于定义细长物体（例如引线框架的引线上的可行接合点的轨迹）的可行点的轨迹的方法和装置。 轨迹引导引线接合器的接合点定位，使得不可接受的引线键合的数量减少。 轨迹是适于接收键的区域，并且被定义为使得对于以其为中心的任何键，形成不可接受的键的概率通过将引线边缘的至少一部分偏离最小值 偏移距离 最小偏移距离基于各种因素，如用户容差和粘合宽度。 此外，轨迹的定义与铅的形状无关，因此可以使用各种铅的形状。 该轨迹还可以具有用于单个引线的各种形状，并且形状可以根据应用而变化。 本文公开的方法和设备描述了在训练期间检测引线并定义轨迹的大小，位置和形状的各种方法。 优选地，也检测引线尖端，使得可以基于与引线的前部的偏移来定义轨迹。 在操作期间，使用轨迹和/或靶键点的特征来重新定位引线并在键合之前精炼目标键/点。 公开的替代实施例包括检查引线。

公开(公告)号：US08005289B2

公开(公告)日：2011-08-23

申请号：US12611886

申请日：2009-11-03

Applicant: Paul Dutta-Choudhury

Inventor： Paul Dutta-Choudhury

IPC: G06K9/00

CPC classification number: G06K9/00134

Abstract: Methods, computer software, and systems for analyzing a biological specimen (e.g., a cytological specimen) on a slide are provided. Magnified image data frames of the biological specimen are taken at different locations on the slide. An object that is not entirely contained within at least one of the image data frames is identified, and complementary portions of the object respectively located in different ones of the image data frames are matched. A fully integrated object is created from the matched object portions. Attributes of the integrated object are then analyzed.

Abstract translation: 提供了用于分析幻灯片上的生物样本（例如，细胞学标本）的方法，计算机软件和系统。 在幻灯片上的不同位置拍摄生物样本的放大图像数据帧。 识别不完全包含在至少一个图像数据帧内的对象，并且分别位于不同图像数据帧中的对象的互补部分相匹配。 从匹配的对象部分创建完全集成的对象。 然后分析集成对象的属性。

公开(公告)号：US20070133856A1

公开(公告)日：2007-06-14

申请号：US11299046

申请日：2005-12-09

Applicant: Paul Dutta-Choudhury

Inventor： Paul Dutta-Choudhury

IPC: G06K9/00

CPC classification number: G06K9/00134

Abstract: Methods, computer software, and systems for analyzing a biological specimen (e.g., a cytological specimen) on a slide is provided. Magnified image data frames of the biological specimen are taken at different locations on the slide. An object that is not entirely contained within at least one of the image data frames is identified, and complementary portions of the object respectively located in different ones of the image data frames are matched. A fully integrated object is created from the matched object portions. Attributes of the integrated object are then analyzed.

Abstract translation: 提供了用于分析幻灯片上的生物样本（例如，细胞学标本）的方法，计算机软件和系统。 在幻灯片上的不同位置拍摄生物样本的放大图像数据帧。 识别不完全包含在至少一个图像数据帧内的对象，并且分别位于不同图像数据帧中的对象的互补部分相匹配。 从匹配的对象部分创建完全集成的对象。 然后分析集成对象的属性。

公开(公告)号：US06920241B1

公开(公告)日：2005-07-19

申请号：US09672493

申请日：2000-09-29

Applicant: Paul Dutta-Choudhury ,  Bradford Safford ,  Joseph Scola

Inventor： Paul Dutta-Choudhury ,  Bradford Safford ,  Joseph Scola

IPC: G06K9/00 ,  G06T7/00 ,  G09G3/34

CPC classification number: G06T7/001

Abstract: A system and method for object inspection incorporates a level of hierarchy so that a single, primary alignment may be performed to generate a pose for multiple inspection regions. As a result, a Training mode may include the specification of a single alignment model window and a list of associated inspection region windows. Similarly, a Run-time Inspection Mode may also include specification of a single alignment model window and a list of associated inspection region windows. Such an implementation addresses the long standing problems associated with performing one or more types of inspection operations in one or more regions of a given sample-object image. By virtue of the hierarchical structure, operation of multiple inspection tools may be performed simultaneously or in any order and inspection of multiple inspection regions may be performed simultaneously or in any order.

Abstract translation: 用于对象检查的系统和方法包括层次级别，使得可以执行单个主要对准以产生多个检查区域的姿势。 结果，训练模式可以包括单个对齐模型窗口的指定和相关联的检查区域窗口的列表。 类似地，运行时检查模式还可以包括单个对齐模型窗口的指定和相关联的检查区域窗口的列表。 这样的实现解决了在给定样本对象图像的一个或多个区域中执行一种或多种类型的检查操作相关联的长期存在的问题。 通过层次结构，可以同时或以任何顺序执行多个检查工具的操作，并且可以同时或以任何顺序执行多个检查区域的检查。

公开(公告)号：US06901171B1

公开(公告)日：2005-05-31

申请号：US09449235

申请日：1999-11-24

Applicant: Paul Dutta-Choudhury ,  Len L. Picard

Inventor： Paul Dutta-Choudhury ,  Len L. Picard

IPC: G06K9/40 ,  G06T5/00

CPC classification number: G06T7/12 ,  G06T2207/20061

Abstract: Methods and apparatuses are disclosed for refining groupings of edge points that represent contours in an image. The methods and apparatuses decrease data dispersion and data quantization effects. The methods and apparatuses are particularly useful for accurate and robust detection of straight line-segment features contained in noisy, cluttered imagery occurring in industrial machine vision applications. Additionally, a measurement criterion of the quality of the detected line segments is introduced.

Abstract translation: 公开了用于精细化表示图像中的轮廓的边缘点的分组的方法和装置。 方法和装置降低数据色散和数据量化效应。 这些方法和装置对于在工业机器视觉应用中发生的嘈杂，杂乱的图像中包含的直线段特征的精确和鲁棒检测特别有用。 另外，引入检测线段的质量的测量标准。

公开(公告)号：US06718074B1

公开(公告)日：2004-04-06

申请号：US09585633

申请日：2000-06-02

Applicant: Paul Dutta-Choudhury ,  Joe Scola ,  Brad Safford ,  Igor Reyzin

Inventor： Paul Dutta-Choudhury ,  Joe Scola ,  Brad Safford ,  Igor Reyzin

IPC: G06K900

CPC classification number: G06T7/0012 ,  G06T3/403 ,  G06T7/0004

Abstract: A machine vision inspection system and method that increase the resolution of object image-data without physically manipulating the physical optics of cameras used during inspection. The exemplary embodiment of the invention provides a system and method that perform what may be characterized as a digital zoom. Using this digital zoom, the resolution of object image-data provided by a camera may be artificially increased. Moreover, in accordance with the exemplary embodiment of the invention, a determination may be made regarding whether portions of the object image-data are under-resolved, and based on that determination, a local digital zoom may be performed on that under-resolved portion of object image-data to increase the resolution of that portion of object image-data. Additionally, the resulting portion of the object image-data with increased image resolution may be scaled-back, such that the portion of increased resolution object image-data may be combined with the other object image-data to provide a faithful, scaled representation of the original sample-object and/or sample-object feature.

Abstract translation: 一种机器视觉检查系统和方法，可以在物理上操纵检查期间使用的相机的物理光学元件，增加对象图像数据的分辨率。 本发明的示例性实施例提供了执行可以被表征为数字变焦的系统和方法。 使用该数字变焦，可以人为地增加由相机提供的对象图像数据的分辨率。 此外，根据本发明的示例性实施例，可以确定对象图像数据的部分是否未被分解，并且基于该确定，可以对未解决部分执行局部数字变焦 的对象图像数据，以增加该部分对象图像数据的分辨率。 另外，具有增加的图像分辨率的对象图像数据的所得部分可以缩小，使得增加的分辨率对象图像数据的部分可以与其他对象图像数据组合以提供忠实的，缩放的 原始样本对象和/或样本对象特征。

公开(公告)号：US5796868A

公开(公告)日：1998-08-18

申请号：US579958

申请日：1995-12-28

Applicant: Paul Dutta-Choudhury

Inventor： Paul Dutta-Choudhury

IPC: G06T5/00 ,  G06K9/48 ,  G06K9/00 ,  G06K9/40 ,  G06K9/52

CPC classification number: G06T7/0083 ,  G06T2207/10016 ,  G06T2207/20061 ,  G06T2207/20081

Abstract: A system and method for filtering edge points in an edge image represented by an array of edge image coordinates each having an edge point magnitude. An edge map array is formed of edge map coordinates each assigned one of three edge indicator values indicative of the relationship between the corresponding edge image coordinate edge point magnitude and two preselected edge magnitude thresholds. An output edge point array is formed of output edge point coordinates to indicate those edge image coordinates and corresponding object image pixels that are validated by the system as being associated with likely object edge points. This is accomplished by constructing a first address stack to track memory addresses at which edge map coordinate values are stored and constructing a second address stack to track memory addresses at which output edge point coordinate values are stored as the output edge point array is constructed. Selected output edge point coordinates are assigned with a preselected edge point value by first scanning the edge map array for edge coordinates assigned the first preselected edge indicator value and, using the stacks to keep track of the addresses of these coordinates, neighboring edge map coordinates are examined for any edge map coordinates assigned with either the first or second preselected edge indicator values. The address of edge map coordinates found to be assigned with either the first or second preselected edge indicator values are then tracked by the stacks.

Abstract translation: 一种用于过滤边缘图像中的边缘点的系统和方法，所述边缘图像由每个具有边缘点幅度的边缘图像坐标阵列表示。 边缘地图阵列由边缘地图坐标形成，每个边缘地图坐标分别指示相应的边缘图像坐标边缘点大小和两个预选边缘幅度阈值之间的关系的三个边缘指示符值之一。 输出边缘点阵列由输出边缘点坐标形成，以指示由系统验证为与可能的对象边缘点相关联的那些边缘图像坐标和对应的对象图像像素。 这是通过构建第一地址堆栈来跟踪存储地图坐标值的存储器地址，并且构建第二地址堆栈来跟踪构建输出边缘点阵列时存储输出边缘点坐标值的存储器地址。 所选择的输出边缘点坐标通过首先扫描边缘图阵列以分配第一预先选择的边缘指示符值的边缘坐标分配预选边缘点值，并且使用堆栈来跟踪这些坐标的地址，相邻边缘图坐标是 检查分配有第一或第二预选边缘指示符值的任何边缘地图坐标。 发现被分配有第一或第二预选边缘指示符值的边缘地图坐标的地址然后被堆栈跟踪。