公开(公告)号：US08401283B2

公开(公告)日：2013-03-19

申请号：US12816779

申请日：2010-06-16

申请人： Kohtaro Sabe ,  Atsushi Okubo ,  Kenichi Hidai

发明人： Kohtaro Sabe ,  Atsushi Okubo ,  Kenichi Hidai

IPC分类号： G06K9/00

CPC分类号： G06K9/6256 ,  G06K9/00288 ,  G06K9/4614

摘要： An information processing apparatus includes the following elements. A learning unit is configured to perform Adaptive Boosting Error Correcting Output Coding learning using image feature values of a plurality of sample images each being assigned a class label to generate a multi-class classifier configured to output a multi-dimensional score vector corresponding to an input image. A registration unit is configured to input a register image to the multi-class classifier, and to register a multi-dimensional score vector corresponding to the input register image in association with identification information about the register image. A determination unit is configured to input an identification image to be identified to the multi-class classifier, and to determine a similarity between a multi-dimensional score vector corresponding to the input identification image and the registered multi-dimensional score vector corresponding to the register image.

摘要翻译： 信息处理装置包括以下要素。 学习单元被配置为使用多个样本图像的图像特征值来执行自适应提升误差校正输出编码学习，每个样本图像被分配类别标签，以生成被配置为输出与输入相对应的多维得分向量的多类分类器 图片。 注册单元被配置为将注册图像输入到多类分类器，并且与关于注册图像的识别信息相关联地注册与输入注册图像相对应的多维分数向量。 确定单元被配置为将要识别的识别图像输入到多类分类器，并且确定与输入的标识图像相对应的多维得分向量与对应于寄存器的登记的多维得分向量之间的相似度 图片。

公开(公告)号：US20100329544A1

公开(公告)日：2010-12-30

申请号：US12816779

申请日：2010-06-16

申请人： Kohtaro Sabe ,  Atsushi Okubo ,  Kenichi Hidai

发明人： Kohtaro Sabe ,  Atsushi Okubo ,  Kenichi Hidai

IPC分类号： G06K9/62

CPC分类号： G06K9/6256 ,  G06K9/00288 ,  G06K9/4614

摘要： An information processing apparatus includes the following elements. A learning unit is configured to perform Adaptive Boosting Error Correcting Output Coding learning using image feature values of a plurality of sample images each being assigned a class label to generate a multi-class classifier configured to output a multi-dimensional score vector corresponding to an input image. A registration unit is configured to input a register image to the multi-class classifier, and to register a multi-dimensional score vector corresponding to the input register image in association with identification information about the register image. A determination unit is configured to input an identification image to be identified to the multi-class classifier, and to determine a similarity between a multi-dimensional score vector corresponding to the input identification image and the registered multi-dimensional score vector corresponding to the register image.

摘要翻译： 信息处理装置包括以下要素。 学习单元被配置为使用多个样本图像的图像特征值来执行自适应提升误差校正输出编码学习，每个样本图像被分配类别标签，以生成被配置为输出与输入相对应的多维得分向量的多类分类器 图片。 注册单元被配置为将注册图像输入到多类分类器，并且与关于注册图像的识别信息相关联地注册与输入注册图像相对应的多维分数向量。 确定单元被配置为将要识别的识别图像输入到多类分类器，并且确定与输入的标识图像相对应的多维得分向量与对应于寄存器的登记的多维得分向量之间的相似度 图片。

公开(公告)号：US08571315B2

公开(公告)日：2013-10-29

申请号：US13288231

申请日：2011-11-03

申请人： Kohtaro Sabe ,  Kenichi Hidai ,  Kiyoto Ichikawa

发明人： Kohtaro Sabe ,  Kenichi Hidai ,  Kiyoto Ichikawa

IPC分类号： G06K9/00 ,  G06K9/38 ,  G06K9/62 ,  G06K9/68

CPC分类号： G06K9/4614 ,  G06K9/6256

摘要： An information processing apparatus includes: a distinguishing unit which, by using an ensemble classifier, which includes a plurality of weak classifiers outputting weak hypotheses which indicates whether a predetermined subject is shown in an image in response to inputs of a plurality of features extracted from the image, and a plurality of features extracted from an input image, sequentially integrates the weak hypotheses output by the weak classifiers in regard to the plurality of features and distinguishes whether the predetermined subject is shown in the input image based on the integrated value. The weak classifier classifies each of the plurality of features to one of three or more sub-divisions based on threshold values, calculates sum divisions of the sub-divisions of the plurality of features as whole divisions into which the plurality of features is classified, and outputs, as the weak hypothesis, a reliability degree of the whole divisions.

摘要翻译： 一种信息处理装置，包括：识别单元，其使用整体分类器，其包括输出弱假设的多个弱分类器，所述弱分类器指示响应于从所述图像提取的多个特征的输入，是否在图像中示出预定对象 图像和从输入图像提取的多个特征，顺序地对由弱分类器输出的关于多个特征的弱假设进行积分，并且基于积分值区分输入图像中是否显示预定对象。 所述弱分类器基于阈值将所述多个特征中的每一个分类为三个或更多个子分割中的一个，并且将所述多个特征的子分割的和除作为将所述多个特征分类成的整个分割，以及 作为弱假设的输出是整个分区的可靠性程度。

公开(公告)号：USRE43873E1

公开(公告)日：2012-12-25

申请号：US13208123

申请日：2011-08-11

申请人： Kenichi Hidai ,  Kohtaro Sabe ,  Kenta Kawamoto

发明人： Kenichi Hidai ,  Kohtaro Sabe ,  Kenta Kawamoto

IPC分类号： G06K9/62 ,  G06K9/00

CPC分类号： G06K9/6282 ,  G06K9/00248 ,  G06K9/6256

摘要： An object detecting device for detecting an object in a given gradation image. A scaling section generates scaled images by scaling down a gradation image input from an image output section. A scanning section sequentially manipulates the scaled images and cutting out window images from them and a discriminator judges if each window image is an object or not. The discriminator includes a plurality of weak discriminators that are learned in a group by boosting and an adder for making a weighted majority decision from the outputs of the weak discriminators. Each of the weak discriminators outputs an estimate of the likelihood of a window image to be an object or not by using the difference of the luminance values between two pixels. The discriminator suspends the operation of computing estimates for a window image that is judged to be a non-object, using a threshold value that is learned in advance.

摘要翻译： 一种用于检测给定灰度图像中的物体的物体检测装置。 缩放部分通过缩小从图像输出部分输入的灰度图像来生成缩放图像。 扫描部分顺序地操纵缩放图像并从中切出窗口图像，并且鉴别器判断每个窗口图像是否是对象。 鉴别器包括通过升压在一组中学习的多个弱识别器和用于从弱识别器的输出进行加权多数决定的加法器。 每个弱识别器通过使用两个像素之间的亮度值的差异来输出窗口图像成为对象的可能性的估计。 鉴别器使用预先学习的阈值暂停对被判断为非对象的窗口图像的计算估计的操作。

公开(公告)号：US08290885B2

公开(公告)日：2012-10-16

申请号：US12381499

申请日：2009-03-12

申请人： Kohtaro Sabe ,  Katsuki Minamino ,  Kenta Kawamoto ,  Hirotaka Suzuki ,  Kenichi Hidai

发明人： Kohtaro Sabe ,  Katsuki Minamino ,  Kenta Kawamoto ,  Hirotaka Suzuki ,  Kenichi Hidai

IPC分类号： G06F17/00 ,  G06N5/00

CPC分类号： G05B19/045 ,  G05B13/0265 ,  G05B2219/23288 ,  G06N99/005

摘要： An information processing apparatus includes: model learning means for self-organizing, on the basis of a state transition model having a state and state transition to be learned by using time series data as data in time series, an internal state from an observation signal obtained by a sensor; and controller learning means for performing learning for allocating a controller, which outputs an action, to each of transitions of a state or each of transition destination states in the state transition model indicating the internal state self-organized by the model learning means.

摘要翻译： 信息处理装置包括：用于自组织的模型学习装置，基于具有通过使用时间序列数据作为时间序列数据而被学习的状态和状态转变的状态转换模型，从获得的观察信号获得的内部状态 通过传感器; 以及控制器学习装置，用于在指示由模型学习装置自组织的内部状态的状态转换模型中执行学习以分配向控制器输出动作的状态或转换目的地状态中的每一个的转换。

公开(公告)号：US07657085B2

公开(公告)日：2010-02-02

申请号：US11089229

申请日：2005-03-24

申请人： Kohtaro Sabe ,  Kenichi Hidai

发明人： Kohtaro Sabe ,  Kenichi Hidai

IPC分类号： G06K9/62 ,  G05B13/02

CPC分类号： G06K9/00248 ,  G06K9/6217 ,  G06K9/6256 ,  G06K9/6282

摘要： In an information processing apparatus, such as an image processing apparatus, positive samples and negative samples are learned by a number of weak classifiers. During learning by the weak classifiers, a value of weighted majority is calculated as a sum of products of the results of classification by the respective weak classifiers and associated weights, and a learning threshold is also calculated. When the number of negative samples is greater than or equal to one half of the number of positive samples, negative samples for which the value of weighted majority is less than the learning threshold are removed.

摘要翻译： 在诸如图像处理装置的信息处理装置中，由许多弱分类器学习正样本和负样本。 在弱分类器的学习期间，加权多数的值被计算为相应弱分类器和关联权重的分类结果的乘积之和，并且还计算学习阈值。 当阴性样本的数量大于或等于阳性样本数量的一半时，取消加权多数值小于学习阈值的负样本。

公开(公告)号：US07630525B2

公开(公告)日：2009-12-08

申请号：US11089932

申请日：2005-03-25

申请人： Kohtaro Sabe ,  Kenichi Hidai

发明人： Kohtaro Sabe ,  Kenichi Hidai

IPC分类号： G06K9/00 ,  G06K9/62

CPC分类号： G06K9/6256 ,  G06K9/00248 ,  G06K9/6282

摘要： In an information processing apparatus, such as a robot that discriminates human faces, nodes are hierarchically arranged in a tree structure. Each of the nodes has a number of weak classifiers. Each terminal node learns face images associated with one label. An upper node learns learning samples of all labels learned by lower nodes. When a window image to be classified is input, discrimination is performed sequentially from upper nodes to lower nodes. When it is determined that the window image does not correspond to a human face, discrimination by lower nodes is not performed, and discrimination proceeds to sibling nodes.

摘要翻译： 在诸如识别人脸的机器人等信息处理装置中，以树结构分层地布置节点。 每个节点都有一些弱分类器。 每个终端节点学习与一个标签相关联的脸部图像。 上级节点学习下层节点学习的所有标签的学习样本。 当输入要分类的窗口图像时，从上层节点到下层节点依次进行鉴别。 当确定窗口图像不对应于人脸时，不执行下位节点的鉴别，并且歧视进行到兄弟节点。

公开(公告)号：US07418312B2

公开(公告)日：2008-08-26

申请号：US10934366

申请日：2004-09-07

申请人： Kenichi Hidai ,  Kohtaro Sabe

发明人： Kenichi Hidai ,  Kohtaro Sabe

IPC分类号： G06F19/00

CPC分类号： B62D57/032

摘要： An object of the present invention is to provide a robot apparatus and a walking control method thereof capable of changing walking control modes in accordance with floor surfaces by discriminating states of the floor surfaces for walking without modifying a step-based walking schedule and capable of providing stable walking even if floor surface states change greatly.A robot apparatus comprises: an action control section 11 to output a walking start instruction; a floor surface discrimination section 12 to discriminate a category for a current floor surface; and a walking control section 13 to compute an adaptive operation amount. The walking control section 13 obtains sensor values of a foot sole sensor and the like from the current floor surface by means of an in-place stepping motion and the like. Based on the sensor value, the walking control section 13 computes the adaptive operation amount as a correction amount from a standard gait model. The floor surface discrimination section 12 performs pattern recognition for the adaptive operation amount to discriminate the category for the current floor surface. The walking control section 13 is supplied with the floor surface category and selects an optimum walking model for the floor surface category. Again from the sensor value, the walking control section 13 computes the adaptive operation amount as a correction amount for the sensor value and provides walking control accordingly.

摘要翻译： 本发明的目的是提供一种机器人装置及其行走控制方法，其能够通过区分用于行走的地板表面的状态来改变与地板表面相应的步行控制模式，而无需修改基于步进的步行时间表，并且能够提供 即使地板表面状态发生很大变化，也能稳定地行走。 机器人装置包括：动作控制部11，输出步行开始指示; 用于区分当前地板表面的类别的地板表面鉴别部分12; 以及步行控制部13，计算自适应动作量。 步行控制部13通过就地步进动作等从当前的地面获得足底传感器等的传感器值。 基于传感器值，步行控制部13根据标准步态模型计算自适应操作量作为校正量。 地板面判别部12对自适应操作量进行模式识别，以区分当前地面的种类。 行走控制部13被提供有地板表面类别，并且选择用于地板表面类别的最佳步行模型。 从传感器值再次，步行控制部13计算自适应操作量作为传感器值的校正量，并相应地提供步行控制。

公开(公告)号：US20060190156A1

公开(公告)日：2006-08-24

申请号：US11347227

申请日：2006-02-06

申请人： Kenichi Hidai ,  Kohtaro Sabe

发明人： Kenichi Hidai ,  Kohtaro Sabe

IPC分类号： G06F17/00

CPC分类号： G05B13/0265

摘要： A learning control apparatus for an autonomous agent including a functional module having a function of multiple inputs and multiple outputs, the function receiving at least one variable and outputting at least one value, includes an estimating unit for estimating a causal relationship of at least one variable, a grouping unit for grouping at least one variable into a variable group in accordance with the estimated causal relationship, a determining for determining a behavior variable corresponding to each of the variable groups, and a layering unit for layering, in accordance with the variable group and the behavior variable, the function corresponding to each variable group, the function receiving the variable grouped into the variable group and outputting the behavior variable.

摘要翻译： 一种用于包括具有多个输入和多个输出的功能的功能模块的自主代理的学习控制装置，所述功能接收至少一个变量并输出至少一个值，所述学习控制装置包括：估计单元，用于估计至少一个变量 ，分组单元，用于根据估计的因果关系将至少一个变量分组成变量组，确定用于确定与每个可变组相对应的行为变量，以及分层单元，用于根据变量组 和行为变量，对应于每个变量组的函数，接收变量的函数分组到变量组中并输出行为变量。

公开(公告)号：US20050280809A1

公开(公告)日：2005-12-22

申请号：US10994942

申请日：2004-11-22

申请人： Kenichi Hidai ,  Kohtaro Sabe ,  Kenta Kawamoto

发明人： Kenichi Hidai ,  Kohtaro Sabe ,  Kenta Kawamoto

IPC分类号： G06T1/00 ,  G06K9/00 ,  G06K9/62 ,  G06K9/68 ,  G06N3/08 ,  G06T7/00 ,  G01N21/00

CPC分类号： G06K9/6282 ,  G06K9/00248 ,  G06K9/6256

摘要： An object detecting device 1 comprises a scaling section 3 for generating scaled images by scaling down a gradation image input from an image output section 2, a scanning section 4 for sequentially manipulating the scaled images and cutting out window images from them and a discriminator 5 for judging if each window image is an object or not. The discriminator 5 includes a plurality of weak discriminators that are learnt in a group by boosting and an adder for making a weighted majority decision from the outputs of the weak discriminators. Each of the weak discriminators outputs an estimate telling the likelihood of a window image to be an object or not by using the difference of the luminance values of two pixels. The discriminator 5 suspends the operation of computing estimates for a window image that is judged to be a non-object, using a threshold value that is learnt in advance.

摘要翻译： 对象检测装置1包括：缩放部分3，用于通过缩小从图像输出部分2输入的灰度图像来生成缩放图像;扫描部分4，用于顺序地操纵缩放图像并从中切出窗口图像;以及鉴别器5， 判断每个窗口图像是否为一个对象。 鉴别器5包括通过升压在一组中学习的多个弱识别器和用于从弱识别器的输出进行加权多数决定的加法器。 每个弱识别器通过使用两个像素的亮度值的差异来输出估计窗口图像成为对象的可能性的估计。 鉴别器5使用预先学习的阈值来暂停对被判断为非对象的窗口图像的计算估计的操作。