专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US5751910A Neural network solder paste inspection system 失效
标题翻译：神经网络焊膏检测系统
公开(公告)号：US5751910A
公开(公告)日：1998-05-12
申请号：US447282
申请日：1995-05-22
申请人： Steven M. Bryant , Kenneth H. Loewenthal
发明人： Steven M. Bryant , Kenneth H. Loewenthal
IPC分类号： G06F15/18 , G05B13/02 , G06N3/00 , G06N3/04 , H05K3/34 , G06K9/00
CPC分类号： G06N3/0436 , Y10S706/912
摘要： A solder paste brick inspection and physical quality scoring system 10 employs a neural network 70 trained with a fuzzified output vector. An image of solder paste bricks 64 on a printed circuit board 12 is acquired by a CCD camera 30. Values of a predetermined set of brick metrics are extracted from the image by a computer 28 and used as a crisp input vector to trained neural network 70. A defuzzifier 76 converts a fuzzy output vector from neural network 70 into a crisp quality score output which can be used for monitoring and process control.
摘要翻译：焊膏砖检查和物理质量评分系统10采用用模糊化输出向量训练的神经网络70。通过CCD摄像机30获取印刷电路板12上的焊膏砖64的图像。计算机28从图像中提取预定的一组砖尺度的值，并将其用作训练神经网络70的清晰输入向量解模糊器76将来自神经网络70的模糊输出向量转换成可用于监视和过程控制的清晰质量得分输出。

2. 发明授权

US5712922A Neural network optical character recognition system and method for classifying characters in a moving web 失效
标题翻译：神经网络光学字符识别系统和移动网络中的字符分类方法
公开(公告)号：US5712922A
公开(公告)日：1998-01-27
申请号：US340270
申请日：1994-11-15
申请人： Kenneth H. Loewenthal , Steven M. Bryant
发明人： Kenneth H. Loewenthal , Steven M. Bryant
IPC分类号： G06F15/18 , G06K9/20 , G06K9/34 , G06K9/66 , G06N3/00 , G06N99/00 , G06K9/62 , G06F7/38
CPC分类号： G06K9/2036 , G06K9/66 , G06K2209/01 , Y10S706/904
摘要： A neural network based optical character recognition technique is presented for identifying characters in a moving web. Image acquisition means defines an imaging window through which the moving web passes such that the characters printed thereon can be imaged. Classification data is extracted and accumulated for each printed web character passing through the imaging window. A light source provides transmissive illumination of the web as it is being imaged. A neural network accelerator is coupled to the image acquisition means for intelligent processing of the accumulated classification data to produce therefrom printed character classification information indicative of each corresponding character imaged. A processor is coupled to the accelerator for converting the classification information into the appropriate ASCII character code. The technique is particularly useful for reading dot-matrix-type characters on a noisy, semi-transparent background at fast real-time rates. A neural network algorithm based recognition method is also described.
摘要翻译：提出了一种基于神经网络的光学字符识别技术，用于识别移动网络中的字符。图像获取装置定义一个成像窗口，移动的网通过该窗口，使得印刷在其上的字符能被成像。为通过成像窗口的每个打印的网页字符提取和累积分类数据。光源在成像时提供幅材的透射照明。神经网络加速器耦合到图像获取装置，用于对累积的分类数据进行智能处理，从而产生指示每个相应字符被成像的印刷字符分类信息。处理器耦合到加速器，用于将分类信息转换成适当的ASCII字符代码。该技术对于以高速实时速率在嘈杂，半透明背景上读取点阵型字符特别有用。还描述了基于神经网络算法的识别方法。

3. 发明授权

US5586223A High speed segmented neural network and fabrication method 失效
标题翻译：高速分段神经网络及其制作方法
公开(公告)号：US5586223A
公开(公告)日：1996-12-17
申请号：US389550
申请日：1995-02-15
申请人： Steven M. Bryant , Kenneth H. Loewenthal
发明人： Steven M. Bryant , Kenneth H. Loewenthal
IPC分类号： G06F15/18 , G06G7/60 , G06N3/04 , G06N3/063 , G06N99/00 , G06F15/80
CPC分类号： G06N3/0454 , G06N3/063
摘要： A high speed, feed forward, segmented neural network and fabrication technique are described. The segmented network includes a plurality of network layers stacked in an ascending pyramid fashion. The network layers are structured with a plurality of subnetworks, and within each subnetwork exists a plurality of nodes structured in a fully interconnected and/or partially interconnected layered neural network arrangement. The inputs and outputs of each subnetwork are one bit digital values constrained to `0` or `1`, while any number of nodes with any number of layers may be modeled for each subnetwork. Each subnetwork is independent of all other subnetworks in a given network layer, and thus, each network layer is segmented. In hardware implementation, each subnetwork comprises a simple memory device, such as a RAM or PROM look-up table. The speed of the neural network system is high and largely dictated by the access time of the memory devices used.
摘要翻译：描述了高速，前馈，分段神经网络和制造技术。分段网络包括以上升金字塔方式堆叠的多个网络层。网络层由多个子网构成，并且在每个子网内存在以完全互连和/或部分互连的分层神经网络布置构造的多个节点。每个子网络的输入和输出是被限制为“0”或“1”的一位数字值，而任何数量的具有任何层数的节点可以为每个子网建模。每个子网独立于给定网络层中的所有其他子网，因此每个网络层被分段。在硬件实现中，每个子网包括诸如RAM或PROM查找表之类的简单存储器件。神经网络系统的速度很高，主要由使用的存储器件的访问时间决定。

4. 发明授权

US5294942A Non-impact printer apparatus and method for recording on a moving medium 失效
标题翻译：用于在移动介质上记录的无冲击打印机装置和方法
公开(公告)号：US5294942A
公开(公告)日：1994-03-15
申请号：US733478
申请日：1991-07-22
申请人： Kenneth H. Loewenthal , Steven M. Bryant , Thomas F. Powers
发明人： Kenneth H. Loewenthal , Steven M. Bryant , Thomas F. Powers
IPC分类号： G06K15/12 , H04N1/40 , A41J2/45
CPC分类号： H04N1/4005 , G06K15/1247 , H04N1/40031
摘要： A non-impact printer apparatus and method for recording on a medium that is moving at a variable rate. A plurality of recording elements arranged transverse to movement of the medium records respective pixels. Each pixel is formed by enabling each recording element a plurality of times during a pixel recording period. To maintain uniformity in pixel recording a duration for recording a sub-pixel and the number of sub-pixels per pixel are adjusted in accordance with the speed of the recording medium. The sum of the sub-pixel recording durations during a pixel recording period comprises the pixel exposure time.
摘要翻译：一种用于在以可变速率移动的介质上记录的无冲击打印机装置和方法。与介质的移动横向排列的多个记录元件记录各个像素。每个像素通过在像素记录周期期间使每个记录元件多次来形成。为了保持像素记录中的均匀性，用于记录子像素的持续时间，并且根据记录介质的速度来调整每像素的子像素的数量。在像素记录周期期间的子像素记录持续时间的总和包括像素曝光时间。

5. 发明授权

US07623722B2 Animated display for image manipulation and correction of digital image 有权
标题翻译：用于图像处理和数字图像校正的动画显示
公开(公告)号：US07623722B2
公开(公告)日：2009-11-24
申请号：US10693040
申请日：2003-10-24
申请人： Wayne E. Prentice , Steven M. Bryant , Jorge L. Rivera
发明人： Wayne E. Prentice , Steven M. Bryant , Jorge L. Rivera
IPC分类号： G06K9/40 , G06K9/00 , H04N5/00 , G06F15/00 , G06K1/00 , G06K15/12 , H04N1/00 , H04N1/40 , H04N1/38 , G09G5/02 , G06F3/00
CPC分类号： G06F3/04845
摘要： A method for selectively optimizing a plurality of image characteristics for captured images, that includes modifying two or more one-dimensional image characteristic controls using a single loop position controller having one-dimensional control. The single loop position controller traverses useful ranges of each of the two or more one-dimensional image characteristic controls. Additionally, a user cycles through several combinations of the two or more one-dimensional image characteristic controls within a video loop; and has a means of selecting a desired image rendered according to the two or more one-dimensional image characteristic controls.
摘要翻译：一种用于选择性地优化用于捕获的图像的多个图像特征的方法，其包括使用具有一维控制的单个循环位置控制器修改两个或更多个一维图像特征控制。单循环位置控制器遍历两个或更多个一维图像特征控制中的每一个的有用范围。另外，用户在视频循环内循环两个或多个一维图像特征控制的若干组合; 并且具有选择根据两个或更多个一维图像特征控制呈现的期望图像的手段。

6. 发明申请

US20080008516A1 Printer dock with two position tray 失效
标题翻译：打印机底座带有两个位置托盘
公开(公告)号：US20080008516A1
公开(公告)日：2008-01-10
申请号：US11481302
申请日：2006-07-05
申请人： Steven M. Bryant , Steven S. Chapman , James G. Rydelek
发明人： Steven M. Bryant , Steven S. Chapman , James G. Rydelek
IPC分类号： B41J13/10
CPC分类号： B41J13/103 , B41J3/36 , B41J3/44 , B41J3/445 , B41J13/0081 , H04N1/00278 , H04N2201/001 , H04N2201/0058 , H04N2201/0084
摘要： A printer dock for use with a digital data terminal, has a housing having a paper inlet portal, a first tray receiver, and a second tray receiver. The first tray receiver adjoins the paper inlet portal. The second tray receiver is disposed in spaced relation to the paper inlet portal. A printer engine is disposed in the housing. The printer engine has a paper path extending from an entrance facing the paper inlet portal. A device interface is operatively connectable to the data terminal and the printer engine. A paper tray is alternately mountable in a use position and a storage position. The paper tray, in the use position, is located in the first tray receiver and is disposed in operative relation to the printer engine. The paper tray, in the storage position, is located within the second tray receiver.
摘要翻译：一种与数字数据终端一起使用的多功能底座打印机，具有一个具有纸张进口入口的外壳，一个第一托盘接收器和一个第二托盘接收器。第一个托盘接收器邻近进纸口。第二托盘接收器与纸张入口入口间隔开设置。打印机引擎设置在外壳中。打印机引擎具有从朝向纸入口入口的入口延伸的纸路径。设备接口可操作地连接到数据终端和打印机引擎。纸盘交替安装在使用位置和存放位置。在使用位置的纸盘位于第一托盘接收器中并且与打印机引擎相关地设置。处于存储位置的纸盘位于第二托盘接收器内。

7. 发明申请

US20080166102A1 IMAGE DIGITAL PROCESSING BASED ON EDIT STATUS 失效
标题翻译：基于编辑状态的图像数字处理
公开(公告)号：US20080166102A1
公开(公告)日：2008-07-10
申请号：US11684663
申请日：2007-03-12
申请人： John R. Fredlund , Aaron T. Deever , Steven M. Bryant , Kenneth A. Parulski , Robert J. Parada
发明人： John R. Fredlund , Aaron T. Deever , Steven M. Bryant , Kenneth A. Parulski , Robert J. Parada
IPC分类号： H04N5/93
CPC分类号： H04N5/782 , H04N5/85 , H04N5/907
摘要： In a method for producing output from an image record, the type of input unit supplying the image record is determined. The image record is classified into one of a predetermined plurality of different edit statuses. The classifying utilizes as input the determined type of input unit. The edit statuses are inclusive of unedited status and very-edited status. The image record is digitally enhanced. During the enhancing material alteration of the image record is constrained. The enhancing is more constrained when the edit status is very-edited and less when the edit status is other than very-edited. The image record is rendered to match a predetermined output unit. The resulting image record is output in viewable form.
摘要翻译：在从图像记录产生输出的方法中，确定提供图像记录的输入单元的类型。图像记录被分类为预定的多个不同的编辑状态之一。分类用作输入单元的确定类型。编辑状态包括未编辑的状态和非常编辑的状态。图像记录数字增强。在增强材料变化的图像记录受到限制。当编辑状态被非常编辑时，增强是更受约束的，而编辑状态不是被非常编辑时更少。渲染图像记录以匹配预定的输出单元。生成的图像记录以可见的形式输出。

8. 发明授权

US07349010B2 Digital camera including an on-line sales mode 失效
标题翻译：数码相机包括在线销售模式
公开(公告)号：US07349010B2
公开(公告)日：2008-03-25
申请号：US10412457
申请日：2003-04-11
申请人： Steven M. Bryant , Kenneth A. Parulski , Lonne R. Lyon , Edward O. Wolf
发明人： Steven M. Bryant , Kenneth A. Parulski , Lonne R. Lyon , Edward O. Wolf
IPC分类号： H04N5/222
CPC分类号： H04N1/00185 , H04N1/00196 , H04N1/2112 , H04N5/23293 , H04N2201/0008
摘要： A digital camera comprises an image sensor for capturing a plurality of images; a processor for processing the plurality of images and producing a corresponding plurality of digital image files; a memory for storing the plurality of digital image files; and a mode control for selecting between a normal mode and an on-line sales mode.
摘要翻译：数码相机包括用于捕获多个图像的图像传感器; 用于处理所述多个图像并产生相应的多个数字图像文件的处理器; 用于存储多个数字图像文件的存储器; 以及用于在正常模式和在线销售模式之间进行选择的模式控制。

9. 发明授权

US07034880B1 System and camera for transferring digital images to a service provider 失效
公开(公告)号：US07034880B1
公开(公告)日：2006-04-25
申请号：US09718018
申请日：2000-11-20
申请人： Jay A. Endsley , Steven M. Bryant , John R. Fredlund , Kenneth A. Parulski
发明人： Jay A. Endsley , Steven M. Bryant , John R. Fredlund , Kenneth A. Parulski
IPC分类号： H04N5/222
CPC分类号： H04N1/00132 , H04N1/00137 , H04N1/00145 , H04N1/00148 , H04N1/00164 , H04N1/00172 , H04N1/00188 , H04N1/00204 , H04N1/0044 , H04N1/2112 , H04N2101/00 , H04N2201/0086 , H04N2201/3226 , H04N2201/3242 , H04N2201/3278
摘要： A method and digital camera for capturing images to be provided to a service provider. The camera includes a capture device for capturing digital images; a memory for storing the captured digital images; an interface to a communications device for transmitting the stored images to a service provider; wherein the digital interface automatically transfers images to the network service provider during a specified time period. Rechargeable batteries may be provided for powering the digital camera; wherein the communications device recharges the batteries using the interface. The camera may further include a display for displaying digital files stored in the digital memory and displaying the at least one order status confirmation file on the display.

10. 发明授权

US06462772B1 Method of calibrating image scanning apparatus of a photographic film scanner 失效
标题翻译：摄影胶片扫描仪的图像扫描装置的校准方法
公开(公告)号：US06462772B1
公开(公告)日：2002-10-08
申请号：US09219032
申请日：1998-12-23
申请人： Steven M. Bryant
发明人： Steven M. Bryant
IPC分类号： H04N5253
CPC分类号： H04N1/407 , H04N1/4076
摘要： In a photographic film digital image scanner, long term adverse effects on image scanning system response are compensated for by establishing an initial gain calibration profile in advance of commencing image scanning operation. A reference Dmin value is established at the same time, preferably from film Dmin regions associated with the bar code pattern located in the leader portion of the film. When image scanning operation is commenced, an updated Dmin value is established for the frame being scanned and the initial gain calibration profile is adjusted by multiplying the pixel-by-pixel values in the profile by the ratio of the reference Dmin over the up-dated frame-specific Dmin value. Preferably, the same image sensor photosites used to establish the reference Dmin value are used for establishing the frame-specific Dmin value and the readings are taken from Dmin regions of the film associated the bar code pattern adjacent the frame to be scanned.
摘要翻译：在摄影胶片数字图像扫描仪中，通过在开始图像扫描操作之前建立初始增益校准轮廓来补偿对图像扫描系统响应的长期不利影响。同时建立参考Dmin值，优选地从与位于胶片的引导部分中的条形码图案相关联的胶片Dmin区域。当图像扫描操作开始时，为被扫描的帧建立更新的Dmin值，并且通过将该轮廓中的逐像素值乘以参考Dmin的比率来调整初始增益校准轮廓。帧特定Dmin值。优选地，用于建立参考Dmin值的相同图像传感器光斑用于建立特定于帧的Dmin值，并且从与要扫描的帧相邻的条形码图案的胶片的Dmin区域获取读数。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式