利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

3 条记录 (耗时 0.012 秒)

    Object/anti-object neural network segmentation
    1.
    发明授权
    Object/anti-object neural network segmentation 失效
    对象/反对象神经网络分割

    公开(公告)号：US5245672A

    公开(公告)日：1993-09-14

    申请号：US847490

    申请日：1992-03-09

    申请人： Charles L. Wilson ,  Michael D. Garris ,  Robert A. Wilkinson, Jr.

    发明人： Charles L. Wilson ,  Michael D. Garris ,  Robert A. Wilkinson, Jr.

    IPC分类号： G06K9/20 ,  G06K9/32 ,  G06K9/46 ,  G06K9/62

    CPC分类号： G06K9/6281 ,  G06K9/00463 ,  G06K9/4628 ,  G06K2209/01

    摘要： The system of the present invention applies self-organizing and/or supervd learning network methods to the problem of segmentation. The segmenter receives a visual field, implemented as a sliding window and distinguishes occurrences of complete characters from occurrences of parts of neighboring characters. Images of isolated whole characters are true objects and the opposite of true objects are anti-objects, centered on the space between two characters. The window is moved across a line of text producing a sequence of images and the segmentation system distinguishes true objects from anti-objects. Frames classified as anti-objects demarcate character boundaries, and frames classified as true objects represent detected character images. The system of the present invention may be a feedforward adaption using a symmetric triggering network. Inputs to the network are applied directly to the separate associative memories of the network. The associative memories produce a best match pattern output for each part of the input data. The associative memories provide two or more subnetworks which define data subsets, such as objects or anti-objects, according to previously learned examples. Multi-layer perceptron architecture may also be used in the system of the present invention rather than the symmetrically triggered feedforward adaptation with tradeoffs in training time but advantages in speed.

    摘要翻译： 本发明的系统将自组织和/或监督学习网络方法应用于分割问题。 分割器接收视觉场，实现为滑动窗口，并将完整字符的出现与相邻字符部分的出现区分开。 孤立的整个人物的图像是真实的对象，真实对象的反面是反对象，以两个字符之间的空间为中心。 窗口被移动穿过一行文本，产生一系列图像，分割系统将真实对象与反对象区分开来。 划分为反对象的帧划分字符边界，分类为真实对象的帧表示检测到的字符图像。 本发明的系统可以是使用对称触发网络的前馈适配。 对网络的输入直接应用于网络的单独关联存储器。 关联存储器为输入数据的每个部分产生最佳匹配模式输出。 关联存储器根据先前学习的示例提供两个或更多个子网，其定义数据子集，诸如对象或反对象。 多层感知器架构也可以用于本发明的系统中，而不是在训练时间内具有权衡的对称触发的前馈适配，而在速度方面有优势。

    STANDARD CALIBRATION TARGET FOR CONTACTLESS FINGERPRINT SCANNERS
    3.
    发明申请
    STANDARD CALIBRATION TARGET FOR CONTACTLESS FINGERPRINT SCANNERS 有权
    用于无缝指纹扫描仪的标准校准目标

    公开(公告)号：US20130070977A1

    公开(公告)日：2013-03-21

    申请号：US13623898

    申请日：2012-09-21

    申请人： Shahram Orandi ,  Fred Byers ,  Stephen Harvey ,  Michael D. Garris ,  Stephen S. Wood ,  John M. Libert ,  Jin Chu Wu

    发明人： Shahram Orandi ,  Fred Byers ,  Stephen Harvey ,  Michael D. Garris ,  Stephen S. Wood ,  John M. Libert ,  Jin Chu Wu

    IPC分类号： G06K9/78

    CPC分类号： G06K9/00033 ,  G06K9/00013 ,  G06K9/00073 ,  G06K9/00087 ,  H04N13/246 ,  H04N17/002

    摘要： A contactless, three-dimensional fingerprint scanner apparatus, method, and system are described. The contactless fingerprint scanner can provide either, or both, topographical contrast of three-dimensional fingerprint features and optical contrast of a three-dimensional fingerprint surface. Data captured from scanning of a target with known geometric features mimicking fingerprint features can be examined as images or surface plots and analyzed for fidelity against the known target feature specifications to evaluate or validate device capture performance as well as interoperability. The target can be used by scanner vendors and designers to validate their devices, as well as to perform type certification.

    摘要翻译： 描述了非接触式三维指纹扫描仪装置，方法和系统。 非接触式指纹扫描仪可以提供三维指纹特征的三维指纹表面的三维指纹特征和光学对比度的地形对比度。 可以将具有模拟指纹特征的已知几何特征的目标扫描获得的数据作为图像或曲面图进行检查，并根据已知目标特征规范分析保真度，以评估或验证设备捕获性能以及互操作性。 扫描仪供应商和设计人员可以使用目标来验证其设备，以及执行型式认证。