Pattern recognition scheme using probabilistic models based on mixtures
distribution of discrete distribution
    61.
    发明授权
    Pattern recognition scheme using probabilistic models based on mixtures distribution of discrete distribution 失效
    基于离散分布的混合分布的概率模型的模式识别方案

    公开(公告)号:US6064958A

    公开(公告)日:2000-05-16

    申请号:US934376

    申请日:1997-09-19

    IPC分类号: G06K9/62 G10L15/14 G10L9/06

    CPC分类号: G10L15/144 G06K9/6297

    摘要: A pattern recognition scheme using probabilistic models that are capable of reducing a calculation cost for the output probability while improving a recognition performance even when a number of mixture component distributions of respective states is small, by arranging distributions with low calculation cost and high expressive power as the mixture component distribution. In this pattern recognition scheme, a probability of each probabilistic model expressing features of each recognition category with respect to each input feature vector derived from each input signal is calculated, where the probabilistic model represents a feature parameter subspace in which feature vectors of each recognition category exist and the feature parameter subspace is expressed by using mixture distributions of one-dimensional discrete distributions with arbitrary distribution shapes which are arranged in respective dimensions. Then, a recognition category expressed by a probabilistic model with a highest probability among a plurality of probabilistic models is outputted as a recognition result.

    摘要翻译: 使用概率模型的模式识别方案,即使当各种状态的混合分量分布的数量少时,也能够通过以低的计算成本和高的表现力排列分布来提高识别性能,从而降低输出概率的计算成本, 混合物成分分布。 在该模式识别方案中,计算表示从每个输入信号导出的每个输入特征矢量的每个识别类别的特征的每个概率模型的概率,其中概率模型表示其中每个识别类别的特征向量的特征参数子空间 存在,并且通过使用以相应尺寸排列的具有任意分布形状的一维离散分布的混合分布来表示特征参数子空间。 然后,输出由多个概率模型中具有最高概率的概率模型表示的识别类别作为识别结果。

    Scheme for model adaptation in pattern recognition based on Taylor
expansion
    62.
    发明授权
    Scheme for model adaptation in pattern recognition based on Taylor expansion 失效
    基于泰勒扩展的模式识别模型适应方案

    公开(公告)号:US6026359A

    公开(公告)日:2000-02-15

    申请号:US929879

    申请日:1997-09-15

    摘要: A model adaptation scheme in the pattern recognition, which is capable of realizing a fast, real time model adaptation and improving the recognition performance. This model adaptation scheme determines a change in a parameter expressing a condition of pattern recognition and probabilistic model training between an initial condition at a time of acquiring training data used in obtaining a model parameter of each probabilistic model and a current condition at a time of actual recognition. Then, the probabilistic models are adapted by obtaining a model parameter after a condition change by updating a model parameter before a condition change according to the determined change, when the initial condition and the current condition are mismatching. The adaptation processing uses a Taylor expansion expressing a change in the model parameter in terms of a change in the parameter expressing the condition.

    摘要翻译: 模式识别中的模型适应方案,能够实现快速,实时的模型适应和提高识别性能。 该模型适应方案确定在获取用于获得每个概率模型的模型参数的实际数据与实际时间的当前条件之间的初始条件时,表示模式识别和概率模型训练条件的参数的变化 承认。 然后,当初始条件和当前条件不匹配时,通过根据所确定的变化更新条件改变之前的模型参数,通过在条件改变之后获得模型参数来适应概率模型。 根据表示条件的参数的变化,适应处理使用表示模型参数的变化的泰勒展开。

    Alerting device for use with a pager
    63.
    发明授权
    Alerting device for use with a pager 失效
    用于寻呼机的警报设备

    公开(公告)号:US6014572A

    公开(公告)日:2000-01-11

    申请号:US892906

    申请日:1997-07-15

    申请人: Satoshi Takahashi

    发明人: Satoshi Takahashi

    IPC分类号: H04Q7/14 G08B5/22 H04Q7/20

    CPC分类号: G08B5/229

    摘要: In an alerting device for a pager, when a call with a message is received within a predetermined period of time of reception of an initial reception of a call including a message identical to that of the received message, the device conducts an alerting operation to help the user of the pager easily recognize the call. After the predetermined period of time, the device performs an ordinary notification to thereby minimize the current consumption. When a radio section receives a signal including data which matches an address assigned to the pager, a time counter initiates measuring the predetermined period of time and a decoder stores a message obtained from the signal in an RAM. When a next message is received, the decoder compares the received message with the stored one. When "matching" occurs between the messages, the decoder drives a loudspeaker with a large volume or a vibration motor with an increased amplitude when the time counter is measuring the predetermined period of time. After lapse of the predetermined period of time, the decoder drives the loudspeaker with an ordinary volume or a vibration motor with a usual amplitude.

    摘要翻译: 在寻呼机的告警装置中,当在接收到包括与接收到的消息的消息相同的消息的呼叫的初始接收的预定时间段内接收到具有消息的呼叫时,该设备进行警报操作以帮助 寻呼机的用户很容易识别呼叫。 在预定时间段之后,设备执行普通通知,从而使电流消耗最小化。 当无线电部分接收到包括与分配给寻呼机的地址匹配的数据的信号时,时间计数器启动测量预定时间段,并且解码器将从该信号获得的消息存储在RAM中。 当接收到下一个消息时,解码器将接收到的消息与存储的消息进行比较。 当消息之间发生“匹配”时,当时间计数器测量预定时间段时,解码器驱动具有大体积的扬声器或具有增加的振幅的振动电机。 在经过预定时间段之后,解码器以普通音量驱动扬声器或具有通常振幅的振动马达。

    Nonvolatile semiconductor memory device and method of reproducing data
of nonvolatile semiconductor memory device
    64.
    发明授权
    Nonvolatile semiconductor memory device and method of reproducing data of nonvolatile semiconductor memory device 有权
    非易失性半导体存储器件和再现非易失性半导体存储器件的数据的方法

    公开(公告)号:US5999448A

    公开(公告)日:1999-12-07

    申请号:US223281

    申请日:1998-12-30

    摘要: There is provided a nonvolatile semiconductor memory device which has a floating gate electrode and writes data by injecting electrons into the floating gate electrode by applying a voltage to a control gate electrode and erases the data by extracting the electrons from the floating gate electrode. The nonvolatile semiconductor memory device which includes a plurality of word lines, a plurality of bit lines intersecting with the plurality of word lines, and a plurality of memory elements each of which is connected to a word line and a bit line at a location where the word line and the bit line are intersected with each other, comprises at least one monitor bit line which intersects with the word lines, and a plurality of monitor elements which are connected to the monitor bit line and the plurality of word lines at locations where the monitor bit line and the plurality of word lines are intersected with each other.

    摘要翻译: 提供了一种非易失性半导体存储器件,其具有浮置栅极电极,并通过向控制栅电极施加电压并将电子注入到浮置栅极电极中来写入数据,并通过从浮置栅电极提取电子来擦除数据。 所述非易失性半导体存储器件包括多个字线,与所述多个字线相交的多个位线,以及多个存储元件,每个所述存储元件连接到字线和位线, 字线和位线彼此相交,包括与字线相交的至少一个监视位线,以及多个监视元件,其连接到监视器位线,并且多个字线在 监视位线和多个字线彼此相交。

    Capillary array electrophoresis system
    65.
    发明授权
    Capillary array electrophoresis system 失效
    毛细管阵列电泳系统

    公开(公告)号:US5938908A

    公开(公告)日:1999-08-17

    申请号:US847735

    申请日:1997-04-22

    摘要: A capillary electrophoresis system comprising a plurality of capillaries filled with a migration medium in which samples to which fluorephore labels are added migrate, a light source providing an excitation light exciting the fluorephore labels, a photo detector detecting a fluorescence radiated from the fluorephore labels, and light focusing means placed between the plurality of capillaries to which the excitation light is irradiated. The parts of the plurality of capillaries to which the excitation light is irradiated and the light focusing means are arranged in a plane shape. The excitation light is irradiated through the plurality of capillaries and the light focusing means, the capillaries and light focusing means alternatively placed in the parts where an excitation light is irradiated. The light focusing means consisting of cylindrical rod lenses. The axis of the cylindrical rod lenses is placed substantially parallel with the capillaries, which can detect the fluorescence from samples by irradiating a plurality of capillaries substantially at the same time in a batch without mechanical scanning of a plurality of capillaries or without optical scanning of a light beam, providing a highly sensitive detection of samples.

    摘要翻译: 一种毛细管电泳系统,包括填充有迁移介质的多个毛细管,其中添加有荧光标记的样品迁移,提供激发荧光标记的激发光的光源,检测从荧光标记物辐射的荧光的光电检测器,以及 放置在激发光照射到的多个毛细管之间的光聚焦装置。 照射激发光的多个毛细管的部分和光聚焦装置被布置成平面形状。 激发光通过多个毛细管和光聚焦装置照射,毛细管和光聚焦装置交替地放置在照射激发光的部分中。 光聚焦装置由圆柱形棒状透镜组成。 圆柱形棒状透镜的轴线基本上与毛细管平行放置,其可以通过基本上同时照射多个毛细管而检测来自样品的荧光,而不需要对多个毛细管进行机械扫描,或者不进行光学扫描 光束,提供高灵敏度的样品检测。

    Wireless computer network communication system and method having at
least two groups of wireless terminals
    66.
    发明授权
    Wireless computer network communication system and method having at least two groups of wireless terminals 失效
    无线计算机网络通信系统和方法具有至少两组无线终端

    公开(公告)号:US5745483A

    公开(公告)日:1998-04-28

    申请号:US536248

    申请日:1995-09-29

    摘要: A wireless computer network communication system for use in an environment in which plural groups can perform network communication at the same time. A channel controller receives a requirement of starting network communication from a computer, controls a wireless channel sensor, and detects a received signal. The wireless channel detector detects which wireless channel is employed by one or more other networks and the channel controller determines the wireless channel to be employed by its own group and also controls the transceiver. The computer starts the network communication through a network controller. The network controller controls the transceiver for transmitting and receiving data by use of the wireless channel, and the same further controls the transmitting/receiving of the requirements and the data from the computer.

    摘要翻译: 一种在多个组可以同时进行网络通信的环境中使用的无线计算机网络通信系统。 信道控制器接收从计算机开始网络通信的需求,控制无线信道传感器,并检测接收到的信号。 无线信道检测器检测一个或多个其他网络使用哪个无线信道,并且信道控制器确定要由其自己的组采用的无线信道,并且还控制收发信机。 计算机通过网络控制器启动网络通信。 网络控制器通过使用无线信道控制收发信机发送和接收数据,同时进一步控制来自计算机的要求和数据的发送/接收。

    Capillary array electrophoresis system
    67.
    发明授权
    Capillary array electrophoresis system 失效
    毛细管阵列电泳系统

    公开(公告)号:US5730850A

    公开(公告)日:1998-03-24

    申请号:US588776

    申请日:1996-01-19

    摘要: A capillary array electrophoresis system by which measurement is conducted using a large number of capillaries. The electrophoresis system includes a plurality of capillary array sheets stacked one on top of another wherein end portions of the capillaries at a detection region are arranged two-dimensionally in such a manner as to elute two-dimensionally a sample from the distal end of each capillary. Excitation light is applied to the sample eluted into a buffer solution, and a two-dimensional fluorescent image is picked up by a detector.

    摘要翻译: 使用大量毛细管进行测量的毛细管阵列电泳系统。 所述电泳系统包括多个毛细管阵列片,所述多个毛细管阵列片层叠在另一个之上,其中毛细管在检测区域的端部以二维排列,从每个毛细管的末端二维地抽出样品 。 将激发光施加到洗脱到缓冲溶液中的样品上,并通过检测器拾取二维荧光图像。

    Workstation for displaying dynamic image with real-time special effects
    68.
    发明授权
    Workstation for displaying dynamic image with real-time special effects 失效
    用于显示具有实时特效的动态图像的工作站

    公开(公告)号:US5694560A

    公开(公告)日:1997-12-02

    申请号:US579131

    申请日:1995-12-27

    摘要: A dynamic-image displaying workstation provided with a display device for displaying an image, a video signal processing circuit for outputting dynamic-image data representing a dynamic image corresponding to a video signal, a first dual port memory for receiving and storing the dynamic-image data outputted from the video signal processing circuit, a second dual port memory for storing data representing pixels of an image to be displayed by the display device, and a third dual port memory for storing data representing a window area. The window area is an area of a window, to which the dynamic-image belongs, and is not covered by any other windows. The workstation is further provided with a fourth dual port memory for storing data representing a dynamic-image display effective area corresponding to the dynamic image represented by the dynamic-image stored in the first dual port memory, a data selector for receiving data read from the first dual port memory at a first input terminal thereof, and data read from the second dual port memory at a second input terminal thereof, for selecting one of the third and fourth dual port memories and receiving data read from the selected dual port memory at a third input terminal thereof, for selecting the data received at the first input terminal thereof if the data read from the third dual port memory is the data representing the window area, and the data read from the third dual port memory is the data representing the dynamic-image display effective area, for selecting the data received at the second input terminal thereof and for outputting a signal representing the selected data to the display device.

    摘要翻译: 一种具有显示图像的显示装置的动态图像显示工作站,用于输出表示与视频信号相对应的动态图像的动态图像数据的视频信号处理电路,用于接收和存储动态图像的第一双端口存储器 从视频信号处理电路输出的数据,用于存储表示要由显示装置显示的图像的像素的数据的第二双端口存储器,以及用于存储表示窗口区域的数据的第三双端口存储器。 窗口区域是动态图像所属的窗口区域,不被任何其他窗口覆盖。 该工作站还设置有第四双端口存储器,用于存储表示与由存储在第一双端口存储器中的动态图像表示的动态图像相对应的动态图像显示有效区域的数据,用于接收从第一双端口存储器读取的数据的数据选择器 在其第一输入端处的第一双端口存储器,以及在其第二输入端从第二双端口存储器读取的数据,用于选择第三和第四双端口存储器中的一个并且接收从所选择的双端口存储器读取的数据 如果从第三双端口存储器读取的数据是表示窗口区域的数据,并且从第三双端口存储器读取的数据是表示动态的数据的数据,则用于选择在其第一输入端子接收的数据 - 图像显示有效区域,用于选择在其第二输入端接收的数据,并将表示所选数据的信号输出到di 喷射装置

    Workstation for simultaneously displaying overlapped windows using a
priority control register
    69.
    发明授权
    Workstation for simultaneously displaying overlapped windows using a priority control register 失效
    工作站,用于使用优先级控制寄存器同时显示重叠的窗口

    公开(公告)号:US5530797A

    公开(公告)日:1996-06-25

    申请号:US43753

    申请日:1993-04-07

    IPC分类号: G09G5/14 G06F12/00

    CPC分类号: G09G5/14 G09G2340/125

    摘要: Pixel data is selected from among first and second dynamic-image memories (DI1, DI2) and a static-memory (SI). In the invention, (a) first and second window area memories (WA1, WA2) for designating shapes and sizes of windows to which video dynamic-images are assigned respectively, (b) first and second dynamic-image area memories (DA1, DA2) for designating memory locations of data stored in both the dynamic-image memories, and (c) a priority control register for designating which video dynamic-image should be displayed in front when video dynamic-images overlap with each other are provided, whereby display for every pixel is executed according to a logical AND value of read-out data from WA1 and read-out data from DA1, a logical AND value of read-out data from WA2 and read-out data from DA2, and read-out data from the priority control register. Thus two video dynamicimages assigned to their respective windows are simultaneously displayed while subjecting them to overlap control, and special techniques including auto-zooming can be accomplished in real time.

    摘要翻译: 从第一和第二动态图像存储器(DI1,DI2)和静态存储器(SI)中选择像素数据。 在本发明中,(a)用于分别指定视频动态图像的窗口的形状和尺寸的第一和第二窗口区域存储器(WA1,WA2),(b)第一和第二动态图像区域存储器(DA1,DA2) ),用于指定存储在动态图像存储器中的数据的存储器位置,以及(c)优先级控制寄存器,用于指定当视频动态图像彼此重叠时应在前面显示哪个视频动态图像,由此显示 对于每个像素,根据来自WA1的读出数据的逻辑与值和来自DA1的读出数据,来自WA2的读出数据的逻辑与值和来自DA2的读出数据和读出数据执行每个像素 从优先级控制寄存器。 因此,分配给它们各自的窗口的两个视频动态图像被同时显示,同时使它们重叠控制,并且可以实时完成包括自动缩放的特殊技术。

    Signal detecting circuit apparatus
    70.
    发明授权
    Signal detecting circuit apparatus 失效
    信号检测电路装置

    公开(公告)号:US5329173A

    公开(公告)日:1994-07-12

    申请号:US37646

    申请日:1993-03-26

    CPC分类号: H03K5/082

    摘要: A signal detecting apparatus is to obtain hysteresis characteristic at a constant ratio regardless of signal amplitude fluctuations by supplying the threshold level adjusting signal corresponding to the input signal amplitude to the comparison circuit. The threshold value setting signal and the hysteresis adjusting signal is supplied to a comparison circuit through an amplifier and a hysteresis adjusting means having the identical characteristic with the peak hold circuit to set a threshold value and a hysteresis range. Therefore even in the case where the operating characteristic of the amplifier and the peak hold circuit for detecting the peak level of signal amplitude fluctuate, the threshold value and the hysteresis range fluctuation corresponding to the fluctuation. As a result, there can not need the adjustment for resetting the operating characteristic, the threshold value and the hysteresis range in each circuit and thus the time and labor required for adjustment can be further saved.

    摘要翻译: 信号检测装置通过将对应于输入信号幅度的阈值电平调整信号提供给比较电路,以恒定比例获得滞后特性,而不管信号振幅波动如何。 阈值设定信号和滞后调整信号通过具有与峰值保持电路具有相同特性的放大器和滞后调整装置提供给比较电路,以设定阈值和滞后范围。 因此,即使在用于检测信号幅度的峰值电平的放大器和峰值保持电路的工作特性波动的情况下,阈值和滞后范围波动对应于波动。 结果,不需要对每个电路中的工作特性,阈值和滞后范围进行复位的调整,因此可以进一步节省调整所需的时间和人力。