会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 2. 发明授权
    • Method and system for extracting features in a pattern recognition system
    • 在模式识别系统中提取特征的方法和系统
    • US06178261B1
    • 2001-01-23
    • US08906253
    • 1997-08-05
    • William J. WilliamsEugene J. ZalubasRobert M. NickelAlfred O. Hero, IIIJeffrey C. O'Neill
    • William J. WilliamsEugene J. ZalubasRobert M. NickelAlfred O. Hero, IIIJeffrey C. O'Neill
    • G06K946
    • G06K9/522G06K2209/01
    • Method and system for extracting features from measurement signals obtained from real world, physical signals by first forming an invariant component of the measurement signals and then using a technique based on a noise subspace algorithm. This technique first casts or projects the transformed measurement signals into separate subspaces for each extraneous variation or group of variations. The subspaces have minimal over-lap. The recognition of a particular invariant component within a pertinent subspace is then preferably performed using Singular Value Decomposition (SVD) techniques to generate a pattern recognition signal. A series of transformations can be used to form an invariant component called the Scale and Translation Invariant Representation (STIR). In one embodiment, the first step is to form an appropriate time-frequency representation such as the Reduced Interference Distribution (RID) or other distribution whose properties are covariant with translations in time and frequency and changes in scale. A series of additional transformations including a scale transform yield the STIR representation. Features are then extracted from a set of STIR representations taken as examples of the desired signal. The STIR approach removes much of the variation due to translation. In bit-mapped documents, the same translation invariant and scale invariant transformations may be made to regularize characters and words. Also, the same feature selection method functions in an image setting. The method has been found to be particularly useful in word spotting in bitmapped documents corrupted by faxing.
    • 通过首先形成测量信号的不变分量,然后使用基于噪声子空间算法的技术,从从真实世界获得的测量信号中提取特征的方法和系统。 这种技术首先将变换的测量信号投射或投影到每个外来变化或变体组的单独的子空间中。 子空间最小化。 然后优选使用奇异值分解(SVD)技术来执行相关子空间内的特定不变分量的识别,以产生模式识别信号。 可以使用一系列变换来形成称为缩放和平移不变量表示(STIR)的不变组件。 在一个实施例中,第一步是形成适当的时间 - 频率表示,例如减少干扰分布(RID)或其属性与时间和频率上的翻译以及尺度变化共同变化的其他分布。 包括尺度变换的一系列附加变换产生STIR表示。 然后从作为期望信号的示例的一组STIR表示中提取特征。 STIR方法消除了由于翻译造成的大部分变化。 在位映射文档中,可以进行相同的转换不变和尺度不变变换以使字符和单词正规化。 此外,相同的特征选择方法在图像设置中起作用。 已经发现该方法对于通过传真损坏的位图文件的单词发现特别有用。
    • 10. 发明授权
    • System for detecting reduced interference time-frequency distribution
    • 用于检测干扰时频分布的系统
    • US5412589A
    • 1995-05-02
    • US899224
    • 1992-06-16
    • William J. WilliamsJechang Jeong
    • William J. WilliamsJechang Jeong
    • G06F17/00G06F17/14G06F17/15G06F15/20
    • G06F17/141
    • A system including method and apparatus for determining the energy content in a signal to be analyzed employs, in an apparatus aspect thereof, a sampling element for producing samples of the electrical signal to be analyzed, a local correlator for producing a plurality of signals corresponding to the samples delayed and multiplicatively combined with their complex conjugates, a bank of delay and adder systems for correlating the outputs of the multipliers in accordance with a binomial kernel, and a Fourier transformer for producing the distribution as a function of time and frequency. Local correlation is effected using a multi-element delay system, such as a shift register. Binomial correlation in the delay and adder systems is achieved by combining additively given samples with samples which are in other states of delay. Moreover, the banks have respectively different numbers of such delay and adder systems. The resulting analytic signal corresponds to a binomial kernel.
    • 一种包括用于确定要分析的信号中的能量含量的方法和装置的系统在其装置方面采用用于产生要分析的电信号样本的采样元件,用于产生对应于多个信号的多个信号的局部相关器 样本与其复共轭延迟和乘法组合,一组延迟和加法器系统,用于根据二项式核对乘法器的输出进行相关,以及用于产生作为时间和频率的函数的分布的傅里叶变换器。 使用诸如移位寄存器的多元件延迟系统来实现局部相关。 在延迟和加法器系统中的二项相关性是通过将加法给定样本与处于其他延迟状态的样本组合来实现的。 此外,银行分别有不同数量的延迟和加法器系统。 所得到的分析信号对应于二项式核。