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
    • 67. 发明授权
    • Texture discrimination method
    • 纹理鉴别法
    • US5448651A
    • 1995-09-05
    • US267552
    • 1994-06-21
    • Hiroshi SakouHitoshi Matsushima
    • Hiroshi SakouHitoshi Matsushima
    • G06K9/38G06T7/40G06K9/34
    • G06K9/38G06K2209/01
    • To reliably discriminate the characters only from the headlines having any background textures, the present invention analyzes the projection profile of a grey image in a headline area in order to automatically set a learning area A in the area of background texture only of the periphery in the headline area, and set a learning area B in the area that includes background texture of a central portion and characters; determines a discriminant function having, as variables, characteristics that use a plurality of pixel densities in the vicinity of a pixel which is processed so that output values of the discriminant function at each of the positions in the learning area A forms a profile with an average Va, that the output values at each of the positions in the learning area B forms a profile with an average Vb, and that the sum of dispersion values of the two profiles becomes smaller than a predetermined value, in order to discriminate in which area is included the pixel which is processed in the headline area; and determines in which area of the background pattern or the characters the pixel is included depending upon whether the output value of the discriminant function is close to the value Va or to the value Vb for each of the pixels in the headline area, in order to discriminate the headline area into areas.
    • 为了从具有任何背景纹理的标题中可靠地区分字符,本发明分析了标题区域中的灰度图像的投影轮廓,以便仅在背景纹理的区域中自动设置学习区域A 标题区域,并在包括中心部分的背景纹理和字符的区域中设置学习区域B; 确定具有在被处理的像素附近使用多个像素密度的特性作为变量的判别函数,使得学习区域A中的每个位置处的判别函数的输出值形成具有平均值的轮廓 Va,学习区域B中的每个位置处的输出值形成具有平均值Vb的轮廓,并且两个轮廓的色散值的和变得小于预定值,以便区分哪个区域是 包括在标题区域中处理的像素; 并且根据判别函数的输出值是否接近于值Va还是标题区域中的每个像素的值Vb,确定背景图案的哪个区域或者像素的包含的字符,以便 将标题区域区分为区域。
    • 68. 发明授权
    • Method of presenting multimedia data in a desired form by comparing and
replacing a user template model with analogous portions of a system
    • 通过将用户模板模型与系统的类似部分进行比较和替换,以所需形式呈现多媒体数据的方法
    • US5278946A
    • 1994-01-11
    • US619506
    • 1990-11-29
    • Shigeru ShimadaHitoshi MatsushimaSeiji KashiokaAkiko Sugihara
    • Shigeru ShimadaHitoshi MatsushimaSeiji KashiokaAkiko Sugihara
    • G06F17/30G06F7/00
    • G06F17/30017Y10S707/99945Y10S707/99948
    • Multimedia data in an object structure include entity objects and relation objects, wherein each entity object has data representing a media element as internal data, and each relation object represents a relation between the entity objects. A relation object is searched according to information specified by a search command, and a searched relation object and internal data of an entity object related thereto are displayed on the screen of a display unit as a set of display elements. To generate a digest of the multimedia data, a matching between a structure portion of a user model and a structure portion of a system model is examined. The user model is generated according to a request of the user and arranged in layers, while the system model has been previously generated. The multimedia data are arranged in layers according to the importance of respective data elements constituting the multimedia data. According to an examination result, a user model structure portion is selectively replaced by a system model structure portion, and the replaced user model is adjusted so as to display the same.
    • 对象结构中的多媒体数据包括实体对象和关系对象,其中每个实体对象具有表示媒体元素的数据作为内部数据,并且每个关系对象表示实体对象之间的关系。 根据搜索命令指定的信息搜索关系对象,并且将搜索到的关系对象和与其相关联的实体对象的内部数据作为一组显示元素显示在显示单元的屏幕上。 为了生成多媒体数据的摘要,检查用户模型的结构部分和系统模型的结构部分之间的匹配。 根据用户的请求生成用户模型,并且在先前已经生成了系统模型的情况下以层为单位来生成用户模型。 多媒体数据根据构成多媒体数据的各个数据元素的重要性分层布置。 根据检查结果,由系统模型结构部分选择性地替换用户模型结构部分,并且调整替换的用户模型以显示它们。
    • 69. 发明授权
    • Multi-layer network and learning method therefor
    • 多层网络及其学习方法
    • US5212767A
    • 1993-05-18
    • US625166
    • 1990-12-10
    • Junichi HigashinoHitoshi Matsushima
    • Junichi HigashinoHitoshi Matsushima
    • G06F15/18G06N3/04G06N3/063G06N3/08G06N3/10G06N99/00
    • G06N3/10G06N3/063
    • A multi-layer neural network comprising an input layer, a hidden layer and an output layer and a learning method for such a network are disclosed. A processor belonging to the hidden layer stores both the factors of multiplication or weights of link for a successive layer nearer to the input layer and the factors of multiplication or weights of link for a preceding layer nearer to the output layer. Namely, the weight for a certain connection is doubly stored in processors which are at opposite ends of that connection. Upon forward calculation, the access to the weights for the successive layer among the weights stored in the processors of the hidden layer can be made by the processors independently from each other. Similarly, upon backward calculation, the access to weights for the preceding layer can be made by the processors independently from each other.
    • 公开了一种包括输入层,隐藏层和输出层的多层神经网络和用于这种网络的学习方法。 属于隐藏层的处理器存储更接近于输入层的连续层的相乘因子或链接权重以及更靠近输出层的先前层的乘法因子或链路权重。 也就是说,某个连接的权重被双重存储在处于与该连接相对端的处理器中。 在正向计算时,可以由处理器彼此独立地对存储在隐藏层的处理器中的权重之间对连续层的权重的访问进行。 类似地,在反向计算时,可以由处理器彼此独立地进行对前一层的权重的访问。
    • 70. 发明授权
    • Motion control apparatus with function to self-form a series of motions
    • 运动控制装置具有自动形成一系列运动的功能
    • US5063492A
    • 1991-11-05
    • US436725
    • 1989-11-15
    • Haruo YodaTakafumi MiyatakeHitoshi Matsushima
    • Haruo YodaTakafumi MiyatakeHitoshi Matsushima
    • G05B13/00B25J9/16B62D57/02G05B13/02G06N3/04G06N3/08
    • G06N3/04B25J9/161B62D57/02
    • A motion control apparatus having a control circuit inputted with a current state of an object to be controlled and outputting a motion quantity, includes a first unit for changing the value of a motion quantity by adding a randomly variable quantity to the motion quantity; a second unit for quantifying, by using an evaluation function, the merit of change in the current state of an object caused by application of the changed motion quantity to the object; a third unit for calculating an optimum motion quantity on the basis of the changed motion quantity if the evaluation function value determined by the second unit is "positive", and on the basis of the changed motion quantity multiplied by -1 if the evaluation function value is "negative"; and a fourth unit for correcting the values of transformation parameters of the control circuit so that an output of the control circuit takes a value nearer to the optimum motion quantity calculated by the third unit than the motion quantity now being outputted from the control circuit.
    • 一种运动控制装置,具有输入被控制对象的当前状态并输出运动量的控制电路,包括:第一单元,用于通过向运动量添加随机可变量来改变运动量的值; 第二单元,用于通过使用评估函数量化由所述改变的运动量应用于所述对象而导致的对象的当前状态的变化的优点; 第三单元,如果由第二单元确定的评估函数值为“正”,则基于改变的运动量计算最佳运动量,并且如果评估函数值 是“负” 以及第四单元,用于校正控制电路的变换参数的值,使得控制电路的输出获得比现在从控制电路输出的运动量更接近由第三单元计算的最佳运动量的值。