公开(公告)号：US08976280B2

公开(公告)日：2015-03-10

申请号：US13824512

申请日：2011-07-06

申请人： Yukio Kamoshida ,  Jun Nyukai ,  Kensuke Habuka ,  Masaki Hiraga

发明人： Yukio Kamoshida ,  Jun Nyukai ,  Kensuke Habuka ,  Masaki Hiraga

IPC分类号： H04N3/14 ,  G06K9/00 ,  H04N5/232

CPC分类号： H04N5/23267 ,  H04N5/2329

摘要： An image processing device includes an image input portion, a motion vector acquisition portion and a distortion component estimation portion. The image input portion inputs a target image which is an image to be processed. The motion vector acquisition portion acquires a motion vector generated by relative movement with respect to a subject drawn in the target image when an imaging device which has generated the target image generates the target image. The distortion component estimation portion estimates a rolling shutter distortion component of the target image based on the motion vector.

摘要翻译： 图像处理装置包括图像输入部分，运动矢量获取部分和失真分量估计部分。 图像输入部输入作为要处理的图像的目标图像。 当生成目标图像的成像装置产生目标图像时，运动矢量获取部分获取相对于在目标图像中绘制的被摄体的相对运动产生的运动矢量。 失真分量估计部分基于运动矢量来估计目标图像的滚动快门失真分量。

公开(公告)号：US08831376B2

公开(公告)日：2014-09-09

申请号：US13541897

申请日：2012-07-05

申请人： Yukio Kamoshida ,  Jun Nyukai ,  Kensuke Habuka ,  Masaki Hiraga

发明人： Yukio Kamoshida ,  Jun Nyukai ,  Kensuke Habuka ,  Masaki Hiraga

IPC分类号： G06K9/40 ,  H04N5/232

CPC分类号： H04N5/23267 ,  H04N5/2329

摘要： An image processing device includes an image input portion, a motion vector acquisition portion and a distortion component estimation portion. The image input portion inputs a plurality of frame images. The motion vector acquisition portion acquires an amount of parallel movement between two frame images. The distortion component estimation portion performs approximation such that a motion matrix indicating a motion between the frame images consists only of a parallel movement component and a rolling shutter distortion component, uses the amount of parallel movement to calculate, for each of the frame images, a tentative distortion coefficient included in the rolling shutter distortion component and uses the plurality of tentative distortion coefficients calculated to estimate a distortion coefficient.

摘要翻译： 图像处理装置包括图像输入部分，运动矢量获取部分和失真分量估计部分。 图像输入部分输入多个帧图像。 运动矢量获取部获取两帧图像之间的平行移动量。 失真分量估计部分执行近似，使得指示帧图像之间的运动的运动矩阵仅由平行运动分量和滚动快门失真分量组成，使用平行运动量针对每个帧图像计算一 包括在滚动快门失真分量中的暂定失真系数，并且使用所计算的多个暂定失真系数来估计失真系数。

公开(公告)号：US20130182135A1

公开(公告)日：2013-07-18

申请号：US13824512

申请日：2011-07-06

申请人： Yukio Kamoshida ,  Jun Nyukai ,  Kensuke Habuka ,  Masaki Hiraga

发明人： Yukio Kamoshida ,  Jun Nyukai ,  Kensuke Habuka ,  Masaki Hiraga

IPC分类号： H04N5/232

CPC分类号： H04N5/23267 ,  H04N5/2329

摘要： An image processing device includes an image input portion, a motion vector acquisition portion and a distortion component estimation portion. The image input portion inputs a target image which is an image to be processed. The motion vector acquisition portion acquires a motion vector generated by relative movement with respect to a subject drawn in the target image when an imaging device which has generated the target image generates the target image. The distortion component estimation portion estimates a rolling shutter distortion component of the target image based on the motion vector.

摘要翻译： 图像处理装置包括图像输入部分，运动矢量获取部分和失真分量估计部分。 图像输入部输入作为要处理的图像的目标图像。 当生成目标图像的成像装置产生目标图像时，运动矢量获取部分获取相对于在目标图像中绘制的被摄体的相对运动产生的运动矢量。 失真分量估计部分基于运动矢量来估计目标图像的滚动快门失真分量。

公开(公告)号：US20070121126A1

公开(公告)日：2007-05-31

申请号：US11473064

申请日：2006-06-23

申请人： Yukio Kamoshida ,  Takahiro Oshino

发明人： Yukio Kamoshida ,  Takahiro Oshino

IPC分类号： G06K15/02

CPC分类号： G06T3/4007 ,  G06T5/002 ,  G06T2200/12

摘要： An image conversion apparatus for geometrically converting input image data and outputting output image data, includes an inverse coordinate conversion unit configured to inversely convert coordinate values on the coordinate system of the output image data into coordinate values on the coordinate system of the input image data based on the geometric conversion. When a region between neighboring pixels of the coordinate system of the input image data is divided into a plurality of divisions, the inverse coordinate conversion unit changes the inversely converted coordinate values so that coordinate values which belong to an identical division of the inversely converted coordinate values have a random two-dimensional array.

摘要翻译： 一种用于对输入图像数据进行几何转换并输出输出图像数据的图像转换装置，包括逆坐标转换单元，被配置为将输出图像数据的坐标系上的坐标值反相转换成基于输入图像数据的坐标系上的坐标值 在几何转换。 当输入图像数据的坐标系的相邻像素之间的区域被划分成多个分区时，逆坐标转换单元改变反转换的坐标值，使得属于逆转换坐标值的相同分割的坐标值 有一个随机的二维数组。

公开(公告)号：US20100063817A1

公开(公告)日：2010-03-11

申请号：US12531219

申请日：2007-03-14

申请人： Soichi Toyama ,  Ikuo Fujita ,  Yukio Kamoshida

发明人： Soichi Toyama ,  Ikuo Fujita ,  Yukio Kamoshida

IPC分类号： G10L15/06

CPC分类号： G10L17/04

摘要： An acoustic model registration apparatus, an talker recognition apparatus, an acoustic model registration method and an acoustic model registration processing program, each of which prevents certainly an acoustic model having a low recognition capability for talker from being registered certainly, are provided.When a talker utters for the N utterances and the utterance sounds of the N utterances are input through the microphone 1, the sound feature quantity extraction part 4 extracts sound feature quantities which indicate the acoustic features of the input utterance sounds, wherein each sound feature quantity has one-to-one correspondence to each utterance, the talker model generation part 5 generates a talker model based on the extracted sound feature quantities for the N utterances, the collation part 6 calculates the degree of individual similarity between the each sound feature quantity of the N utterances and the talker model generated above, and only in the case that all the calculated degrees of similarities of the N utterances are equal to or more than the threshold value, the similarity verifying part 9 directs to register the generated talker model in the talker models' database as a talker model for the talker recognition.

摘要翻译： 提供了一种声学模型登记装置，讲话者识别装置，声学模型登记方法和声学模型登记处理程序，其中，每个都可以防止当前登记具有低的识别能力的声学模型被登记。 当讲话者发出N个话语时，通过麦克风1输入N个发音的发声，声音特征量提取部分4提取指示输入话音的声音特征的声音特征量，其中每个声音特征量 讲话者模型生成部5基于所提取的N个特征量的声音特征量生成说话者模型，计算出每个声音特征量之间的个体相似度的程度， 上面产生的N个话语和说话者模型，并且只有在所有计算出的N个话语的相似程度等于或大于阈值的情况下，相似性验证部分9才指示将所产生的讲话者模型注册在 讲话者模型的数据库作为谈话者识别的谈话者模型。

公开(公告)号：US20050206739A1

公开(公告)日：2005-09-22

申请号：US11083488

申请日：2005-03-18

申请人： Yukio Kamoshida

发明人： Yukio Kamoshida

IPC分类号： H04N5/232 ,  G06T5/00 ,  G06T5/50 ,  G06T7/00 ,  G06T7/20 ,  H04N5/228

CPC分类号： H04N5/23264 ,  G06T5/006 ,  G06T5/50 ,  G06T7/246 ,  H04N5/23248 ,  H04N5/23254 ,  H04N5/23258

摘要： Provided is an image deformation estimating method capable of obtaining information expressing image deformation between frame images caused by a camera shake with almost no influence of photographing conditions. The image deformation estimating method includes: a first step of providing plural local area in plural frame images, which form a moving image, and obtaining plural pieces of first information indicating movements among the plural frame images in the plural local areas, respectively; a second step of obtaining plural pieces of second information, which express image deformation of each of the plural local areas, using the plural pieces of first information; a third step of preparing plural selected rules, selecting the plural pieces of second information for each of the selected rules, and obtaining second representative information, which represents the selected plural pieces of second information, for each of the selected rules; and a fourth step of using the plural pieces of second representative information, which are obtained for the plural selected rules, to obtain third information expressing deformation of an entire image.

摘要翻译： 提供一种图像变形估计方法，其能够几乎不受拍摄条件的影响，获得表示由相机抖动引起的帧图像之间的图像变形的信息。 图像变形估计方法包括：第一步骤，在形成运动图像的多个帧图像中提供多个局部区域，并且分别获得表示多个局部区域中的多个帧图像之间的运动的多条第一信息; 第二步骤，使用多条第一信息获得表示多个局部区域中的每一个的图像变形的多条第二信息; 准备多个所选择的规则的第三步骤，为每个所选择的规则选择多条第二信息，并为每个所选择的规则获得表示所选择的多条第二信息的第二代表信息; 以及第四步骤，使用针对所选择的多个规则获得的多个第二代表信息来获得表示整个图像的变形的第三信息。

公开(公告)号：US07634154B2

公开(公告)日：2009-12-15

申请号：US11473064

申请日：2006-06-23

申请人： Yukio Kamoshida ,  Takahiro Oshino

发明人： Yukio Kamoshida ,  Takahiro Oshino

IPC分类号： G06K9/36

CPC分类号： G06T3/4007 ,  G06T5/002 ,  G06T2200/12

摘要： An image conversion apparatus for geometrically converting input image data and outputting output image data, includes an inverse coordinate conversion unit configured to inversely convert coordinate values on the coordinate system of the output image data into coordinate values on the coordinate system of the input image data based on the geometric conversion. When a region between neighboring pixels of the coordinate system of the input image data is divided into a plurality of divisions, the inverse coordinate conversion unit changes the inversely converted coordinate values so that coordinate values which belong to an identical division of the inversely converted coordinate values have a random two-dimensional array.

摘要翻译： 一种用于对输入图像数据进行几何转换并输出输出图像数据的图像转换装置，包括逆坐标转换单元，被配置为将输出图像数据的坐标系上的坐标值反相转换成基于输入图像数据的坐标系上的坐标值 在几何转换。 当输入图像数据的坐标系的相邻像素之间的区域被划分成多个分区时，逆坐标转换单元改变反转换的坐标值，使得属于逆转换坐标值的相同分割的坐标值 有一个随机的二维数组。

公开(公告)号：US07502052B2

公开(公告)日：2009-03-10

申请号：US11083488

申请日：2005-03-18

申请人： Yukio Kamoshida

发明人： Yukio Kamoshida

IPC分类号： H04N5/228

CPC分类号： H04N5/23264 ,  G06T5/006 ,  G06T5/50 ,  G06T7/246 ,  H04N5/23248 ,  H04N5/23254 ,  H04N5/23258

摘要： Provided is an image deformation estimating method capable of obtaining information expressing image deformation between frame images caused by a camera shake with almost no influence of photographing conditions. The image deformation estimating method includes: a first step of providing plural local area in plural frame images, which form a moving image, and obtaining plural pieces of first information indicating movements among the plural frame images in the plural local areas, respectively; a second step of obtaining plural pieces of second information, which express image deformation of each of the plural local areas, using the plural pieces of first information; a third step of preparing plural selected rules, selecting the plural pieces of second information for each of the selected rules, and obtaining second representative information, which represents the selected plural pieces of second information, for each of the selected rules; and a fourth step of using the plural pieces of second representative information, which are obtained for the plural selected rules, to obtain third information expressing deformation of an entire image.

摘要翻译： 提供一种图像变形估计方法，其能够几乎不受拍摄条件的影响，获得表示由相机抖动引起的帧图像之间的图像变形的信息。 图像变形估计方法包括：第一步骤，在形成运动图像的多个帧图像中提供多个局部区域，并且分别获得表示多个局部区域中的多个帧图像之间的运动的多条第一信息; 第二步骤，使用多条第一信息获得表示多个局部区域中的每一个的图像变形的多条第二信息; 准备多个所选择的规则的第三步骤，为每个所选择的规则选择多条第二信息，并为每个所选择的规则获得表示所选择的多条第二信息的第二代表信息; 以及第四步骤，使用针对所选择的多个规则获得的多个第二代表信息来获得表示整个图像的变形的第三信息。