专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

51. 发明授权

US08027544B2 Information processing method and information processing apparatus 有权
标题翻译：信息处理方法和信息处理装置
公开(公告)号：US08027544B2
公开(公告)日：2011-09-27
申请号：US12120481
申请日：2008-05-14
申请人： Yusuke Mitarai , Masakazu Matsugu , Katsuhiko Mori
发明人： Yusuke Mitarai , Masakazu Matsugu , Katsuhiko Mori
IPC分类号： G06K9/68
CPC分类号： G06K9/6252 , G06K9/6234
摘要： Processing data which belong to different classes and label data indicating the classes to which the processing data belong are input (S20). A distance relationship between the processing data is calculated (S22). An interclass separation degree between the classes is set (S23). The distance relationship is updated based on the label data and the interclass separation degree (S24). A data mapping relationship which approximates the updated distance relationship is calculated (S25).
摘要翻译：输入属于不同类别的处理数据和指示处理数据所属类别的标签数据（S20）。计算处理数据之间的距离关系（S22）。设定类之间的类间分离度（S23）。基于标签数据和间隔间隔度来更新距离关系（S24）。计算近似更新的距离关系的数据映射关系（S25）。

52. 发明申请

US20110206237A1 RECOGNITION APPARATUS AND METHOD THEREOF, AND COMPUTER PROGRAM 有权
标题翻译：识别装置及其方法，以及计算机程序
公开(公告)号：US20110206237A1
公开(公告)日：2011-08-25
申请号：US13030956
申请日：2011-02-18
申请人： Takayuki Saruta , Masakazu Matsugu , Masato Aoba
发明人： Takayuki Saruta , Masakazu Matsugu , Masato Aoba
IPC分类号： G06K9/00
CPC分类号： G06K9/4661 , G01B11/00
摘要： A recognition apparatus for recognizing a position and an orientation of a target object, inputs a captured image of the target object captured by an image capturing apparatus; detects a plurality of feature portions from the captured image, and to extract a plurality of feature amounts indicating image characteristics in each of the plurality of feature portions; inputs property information indicating respective physical properties in the plurality of feature portions on the target object; inputs illumination information indicating an illumination condition at the time of capturing the captured image; determines respective degrees of importance of the plurality of extracted feature amounts based on the respective physical properties indicated by the property information and the illumination condition indicated by the illumination information; and recognizes the position and the orientation of the target object based on the plurality of feature amounts and the respective degrees of importance thereof.
摘要翻译：用于识别目标对象的位置和取向的识别装置输入由图像捕获装置捕获的目标对象的拍摄图像; 从捕获的图像中检测多个特征部分，并且提取指示多个特征部分中的每一个中的图像特征的多个特征量; 输入表示所述目标物体上的所述多个特征部中的各物理属性的属性信息; 输入指示拍摄图像时的照明条件的照明信息; 基于由所述属性信息和由所述照明信息指示的照明条件指示的各个物理属性来确定所述多个提取的特征量的各自的重要程度; 并且基于多个特征量及其各自的重要程度来识别目标对象的位置和取向。

53. 发明授权

US07958070B2 Parameter learning method, parameter learning apparatus, pattern classification method, and pattern classification apparatus 有权
标题翻译：参数学习方法，参数学习装置，模式分类方法和模式分类装置
公开(公告)号：US07958070B2
公开(公告)日：2011-06-07
申请号：US11769143
申请日：2007-06-27
申请人： Yusuke Mitarai , Masakazu Matsugu , Katsuhiko Mori , Kan Torii , Hiroshi Sato
发明人： Yusuke Mitarai , Masakazu Matsugu , Katsuhiko Mori , Kan Torii , Hiroshi Sato
IPC分类号： G06F15/18
CPC分类号： G06K9/00288 , G06K9/00228 , G06K9/00248 , G06K9/6232 , G06K9/6256 , G06K9/6269
摘要： A plurality of pieces of learning data, each associated with a class to which the piece of the learning data belong, are input. In each piece of the learning data, a statistical amount of attribute values of elements in each of specific k parts, k being equal to or larger than 1, is calculated. Each piece of the learning data is mapped in a k-dimensional feature space as a vector having the calculated k statistics amounts as elements. Based on each piece of the mapped learning data and the classes to which the pieces of learning data belong, parameters for classifying input data into one of the plurality of classes are learned in the k-dimensional feature space. By using the parameters, pattern classification can be performed with high speed and high accuracy.
摘要翻译：输入与学习数据所属的类别相关联的多条学习数据。在每个学习数据中，计算k个等于或大于1的特定k个部分中的每个元素的属性值的统计量。每个学习数据被映射到k维特征空间中，作为具有计算的k个统计量的向量作为元素。基于映射的学习数据和学习数据所属的类别，在k维特征空间中学习用于将输入数据分类为多个类中的一个的参数。通过使用参数，可以高速，高精度地执行模式分类。

54. 发明授权

US07912253B2 Object recognition method and apparatus therefor 有权
标题翻译：对象识别方法及其设备
公开(公告)号：US07912253B2
公开(公告)日：2011-03-22
申请号：US11630031
申请日：2005-06-27
申请人： Takashi Suzuki , Katsuhiko Mori , Masakazu Matsugu
发明人： Takashi Suzuki , Katsuhiko Mori , Masakazu Matsugu
IPC分类号： G06K9/00
CPC分类号： G06K9/00281 , G06K9/00275 , G06K9/4628
摘要： A feature extraction unit (23) extracts features of an object of interest included in an image. A local region setting unit (24) sets a local region that includes a feature group required to obtain the shape of the object and a local region that includes a feature group required to obtain the positional relationship. A feature vector generation unit (25) calculates feature vector data in the set local regions. An object recognition unit (26) determines which of different objects the object of interest is, on the basis of the feature vectors.
摘要翻译：特征提取单元（23）提取包括在图像中的感兴趣对象的特征。局部区域设定单元（24）设定包含获取对象形状所需要的特征组的局部区域和包含获取位置关系所需特征组的局部区域。特征向量生成单元（25）计算设定的局部区域中的特征向量数据。基于特征向量，对象识别单元（26）确定感兴趣对象的哪个不同对象。

55. 发明授权

US07894666B2 Image processing method and apparatus, image sensing apparatus, and program 有权
标题翻译：图像处理方法和装置，图像感测装置和程序
公开(公告)号：US07894666B2
公开(公告)日：2011-02-22
申请号：US11571325
申请日：2005-07-27
申请人： Yusuke Mitarai , Masakazu Matsugu , Katsuhiko Mori , Yuji Kaneda
发明人： Yusuke Mitarai , Masakazu Matsugu , Katsuhiko Mori , Yuji Kaneda
IPC分类号： G06K9/00
CPC分类号： G06T7/0012 , G06K9/0061 , G06T7/90 , G06T2207/30041 , G06T2207/30216 , H04N1/624
摘要： An eye region is extracted from an input image. A poor pupil hue quality region is extracted from the eye region. A high brightness region is extracted from the poor pupil hue quality region. A region obtained by excluding the high brightness region from the poor pupil hue quality region is extracted as a red-eye region. The high brightness region is corrected by using a method different from that for the red-eye region.
摘要翻译：从输入图像中提取眼睛区域。从眼部区域提取出不良的瞳孔色调区域。从较差的瞳孔色调质量区域提取高亮度区域。提取通过从较差瞳孔色调质量区域排除高亮度区域而获得的区域作为红眼区域。通过使用与红眼区域不同的方法来校正高亮度区域。

56. 发明申请

US20100328649A1 MEASUREMENT SYSTEM AND MEASUREMENT PROCESSING METHOD 有权
标题翻译：测量系统和测量处理方法
公开(公告)号：US20100328649A1
公开(公告)日：2010-12-30
申请号：US12816069
申请日：2010-06-15
申请人： Kazuyuki Ota , Masakazu Matsugu , Kenji Saitoh
发明人： Kazuyuki Ota , Masakazu Matsugu , Kenji Saitoh
IPC分类号： G01B11/24 , G01B11/30
CPC分类号： G01B11/25
摘要： This invention is directed to extract the scattering characteristic of a measurement target together when measuring the surface shape in a measurement system, which measures the surface shape of a measurement target, by the pattern projection method. To accomplish this, the measurement system includes an illumination unit which irradiates a measurement target with dot pattern light, a reflected light measurement unit which receives the reflected light at a reflection angle almost equal to a incident angle, and a reflected light extraction unit which extracts the inclination of the surface of the measurement target, based on the shift amount between the light receiving position of the received reflected light and a predetermined reference position, and extracts the luminance value of the reflected light and the dot diameter of the dot pattern light as information about the scattering characteristic.
摘要翻译：本发明旨在通过图案投影法在测量测量对象的表面形状的测量系统中测量表面形状时，将测量对象的散射特性一起提取。为了实现这一点，测量系统包括用点阵图形光照射测量对象的照明单元，以几乎等于入射角的反射角接收反射光的反射光测量单元和提取出的反射光提取单元基于接收到的反射光的光接收位置与预定的基准位置之间的偏移量，测量对象的表面的倾斜度，并将点阵图案光的反射光的亮度值和点直径提取为关于散射特性的信息。

57. 发明申请

US20100220338A1 MEASUREMENT APPARATUS AND METHOD FOR MEASURING SURFACE SHAPE AND ROUGHNESS 有权
标题翻译：测量装置和测量表面形状和粗糙度的方法
公开(公告)号：US20100220338A1
公开(公告)日：2010-09-02
申请号：US12669825
申请日：2008-08-14
申请人： Kazuyuki Ota , Hiroshi Yoshikawa , Yusuke Mitarai , Masafumi Takimoto , Kazunori Okudomi , Hiroyuki Shinbata , Kenji Saitoh , Masakazu Matsugu
发明人： Kazuyuki Ota , Hiroshi Yoshikawa , Yusuke Mitarai , Masafumi Takimoto , Kazunori Okudomi , Hiroyuki Shinbata , Kenji Saitoh , Masakazu Matsugu
IPC分类号： G01B11/24
CPC分类号： G01B11/24 , G01B11/30
摘要： A measurement apparatus includes an illumination light setting unit which sets illumination light having an optical characteristic corresponding to a microstructure which is formed on the surface of a measurement target and is to be measured, a measurement unit which measures reflected light when the measurement target is irradiated with the illumination light, and an extraction unit which extracts, from the measured reflected light, information about the surface shape of the measurement target and the microstructure formed on the surface.
摘要翻译：测量装置包括照明光设定单元，其设置具有对应于形成在测量对象物的表面上并要被测量的微结构的光学特性的照明光，测量单元，其在测量对象被照射时测量反射光以及从测量的反射光中提取关于测量对象的表面形状和形成在表面上的微结构的信息的提取单元。

58. 发明授权

US07676441B2 Information processing apparatus, information processing method, pattern recognition apparatus, and pattern recognition method 失效
标题翻译：信息处理装置，信息处理方法，模式识别装置和模式识别方法
公开(公告)号：US07676441B2
公开(公告)日：2010-03-09
申请号：US11149698
申请日：2005-06-10
申请人： Masakazu Matsugu , Katsuhiko Mori , Mie Ishii , Yusuke Mitarai
发明人： Masakazu Matsugu , Katsuhiko Mori , Mie Ishii , Yusuke Mitarai
IPC分类号： G06F15/18
CPC分类号： G06K9/4628 , G06K9/00604 , G06N3/049 , G06N3/08
摘要： In a hierarchical neural network having a module structure, learning necessary for detection of a new feature class is executed by a processing module which has not finished learning yet and includes a plurality of neurons which should learn an unlearned feature class and have an undetermined receptor field structure by presenting a predetermined pattern to a data input layer. Thus, a feature class necessary for subject recognition can be learned automatically and efficiently.
摘要翻译：在具有模块结构的分层神经网络中，由尚未完成学习的处理模块执行用于检测新特征类所需的学习，并且包括多个神经元，其应该学习未知的特征类并具有未确定的受体场通过向数据输入层呈现预定图案来构造。因此，可以自动且有效地学习用于对象识别所需的特征类。

59. 发明申请

US20080253665A1 PATTERN IDENTIFICATION APPARATUS AND METHOD THEREOF, ABNORMAL PATTERN DETECTION APPARATUS AND METHOD THEREOF, AND PROGRAM 有权
标题翻译：模式识别装置及其方法，异常模式检测装置及其方法及程序
公开(公告)号：US20080253665A1
公开(公告)日：2008-10-16
申请号：US12100664
申请日：2008-04-10
申请人： Yusuke Mitarai , Masakazu Matsugu
发明人： Yusuke Mitarai , Masakazu Matsugu
IPC分类号： G06K9/70
CPC分类号： G06K9/6252
摘要： A pattern identification apparatus for identifying one of a plurality of classes defined in advance, to which data of a pattern identification target belongs, comprises a read unit adapted to read out, from a storage unit in correspondence with each of the plurality of classes, a projection rule to a hyperplane which approximates a manifold corresponding to the class in a feature space an input unit adapted to input identification target data; a calculation unit adapted to calculate, for each class, a projection result obtained by projecting the input identification target data to the hyperplane which approximates the manifold corresponding to each of the plurality of classes, on the basis of the projection rule; and an identification unit adapted to identify, on the basis of the projection result of each classes calculated by said calculation unit, one of the plurality of classes to which the identification target data belongs.
摘要翻译：一种模式识别装置，用于识别模式识别目标所属的数据预先定义的多个类别中的一个，包括适于从存储单元中读出与多个类别中的每个类别相对应的读取单元投影规则到近似对应于特征空间中的类别的歧管的超平面，适于输入识别目标数据的输入单元; 计算单元，用于根据所述投影规则，针对每个类计算通过将所述输入的识别目标数据投影到与所述多个类别中的每一个对应的所述歧管的所述超平面上而获得的投影结果; 以及识别单元，其适于基于由所述计算单元计算出的每个类别的投影结果来识别识别目标数据所属的多个类别中的一个。

60. 发明授权

US07437339B2 Pulse signal circuit, parallel processing circuit, pattern recognition system, and image input system 失效
标题翻译：脉冲信号电路，并行处理电路，模式识别系统和图像输入系统
公开(公告)号：US07437339B2
公开(公告)日：2008-10-14
申请号：US11430133
申请日：2006-05-09
申请人： Masakazu Matsugu
发明人： Masakazu Matsugu
IPC分类号： G06N3/00 , G06T1/40
CPC分类号： G06N3/049 , G06K9/00973 , G06K9/4628 , G06N3/063
摘要： A synaptic connection element for connecting neuron elements inputs a plurality of pulsed signals from different neuron elements N1 through N4, effects a common modulation (time window integration or pulse phase/width modulation) on a plurality of predetermined signals among the plurality of pulse signals, and outputs the modulated pulse signals to different signal lines to a neuron element M1. A neural network for representing and processing pattern information by the pulse modulation is thereby downsized in scale.
摘要翻译：用于连接神经元元件的突触连接元件输入来自不同神经元元件N 1至N 4的多个脉冲信号，对多个脉冲中的多个预定信号实现共同调制（时间窗积分或脉冲相位/宽度调制）信号，并将调制脉冲信号输出到不同的信号线到神经元M1。因此，通过脉冲调制来表示和处理图案信息的神经网络的规模缩小。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式