METHOD AND SYSTEM FOR GENERATING A SYMBOL IDENTIFICATION CHALLENGE
    1.
    发明申请
    METHOD AND SYSTEM FOR GENERATING A SYMBOL IDENTIFICATION CHALLENGE 审中-公开
    用于产生符号识别挑战的方法和系统

    公开(公告)号:US20100046790A1

    公开(公告)日:2010-02-25

    申请号:US12196389

    申请日:2008-08-22

    IPC分类号: G06K9/00

    CPC分类号: G06F21/36

    摘要: Provided is a system and method for generating a symbol identification challenge. The method includes receiving a plurality of symbols, e.g. a base symbol and a noise symbol. A first visual property and a second visual property is then applied to each received symbol. Each symbol is then transformed by transitioning the first visual property to the second visual property and the second visual property to the first visual property. A plurality of views is then generated by transitioning between the transformations of each symbol. An associated system for performing the method is also provided.

    摘要翻译: 提供了一种用于生成符号识别挑战的系统和方法。 该方法包括接收多个符号,例如, 基本符号和噪声符号。 然后将第一视觉属性和第二视觉属性应用于每个接收到的符号。 然后通过将第一视觉属性转换为第二视觉属性并将第二视觉特性转换为第一视觉属性来转换每个符号。 然后通过在每个符号的变换之间进行转换来生成多个视图。 还提供了用于执行该方法的关联系统。

    METHOD AND SYSTEM FOR GENERATING A CONTEXTUAL SEGMENTATION CHALLENGE FOR AN AUTOMATED AGENT
    2.
    发明申请
    METHOD AND SYSTEM FOR GENERATING A CONTEXTUAL SEGMENTATION CHALLENGE FOR AN AUTOMATED AGENT 审中-公开
    用于生成自动化代理的上下文分段挑战的方法和系统

    公开(公告)号:US20100302255A1

    公开(公告)日:2010-12-02

    申请号:US12786711

    申请日:2010-05-25

    IPC分类号: G06T15/70

    摘要: Provided is a system and method for generating a contextual segmentation challenge that poses an identification challenge. The method including obtaining at least one ad element and obtaining a test element. The ad element and the test element then combined to provide a composite image. At least one noise characteristic is then applied to the composite image. The composite image is then animated as a plurality of views as a contextual segmentation challenge. A system for performing the method is also provided.

    摘要翻译: 提供了一种用于产生呈现识别挑战的上下文分段挑战的系统和方法。 该方法包括获得至少一个广告元素并获得测试元素。 然后将ad元素和测试元素组合以提供合成图像。 然后将至少一个噪声特性应用于合成图像。 然后,合成图像作为多个视图被动画化为上下文分段挑战。 还提供了一种用于执行该方法的系统。

    METHOD AND SYSTEM FOR GENERATING A REPRESENTATION SYMBOL AS AN IDENTIFICATION CHALLENGE
    3.
    发明申请
    METHOD AND SYSTEM FOR GENERATING A REPRESENTATION SYMBOL AS AN IDENTIFICATION CHALLENGE 审中-公开
    用于产生代表性符号的方法和系统作为识别挑战

    公开(公告)号:US20090297064A1

    公开(公告)日:2009-12-03

    申请号:US12129809

    申请日:2008-05-30

    IPC分类号: G06K9/54

    CPC分类号: G06T11/203

    摘要: Provided is a system and method for generating a representation of a symbol. The method includes receiving a base image symbol. The base image symbol is distinguished into a plurality of initial image components. The initial image components are mapped to a first plane defined by two perpendicular axes, a third axis being normal to the plane. A noise function is selected. A plurality of views are generated, each by applying the selected noise function to each initial image component to provide an adjusted image component, each adjusted image component offset from initial image component. The adjusted image components are collectively orthogonally projected into a viewing plane angled relative to the first plane. The plurality of views are output such as to a database or a display. An associated system for performing the method is also provided.

    摘要翻译: 提供了一种用于生成符号的表示的系统和方法。 该方法包括接收基本图像符号。 基本图像符号被区分为多个初始图像分量。 初始图像分量被映射到由两个垂直轴定义的第一平面,第三轴垂直于平面。 选择噪声功能。 产生多个视图,每个视图通过将选择的噪声函数应用于每个初始图像分量,以提供经调整的图像分量,每个调整图像分量偏离初始图像分量。 经调整的图像分量被共同地投影到相对于第一平面成角度的观察平面中。 多个视图被输出到数据库或显示器。 还提供了用于执行该方法的关联系统。