-
公开(公告)号：US09165343B2
公开(公告)日：2015-10-20
申请号：US14035325
申请日：2013-09-24
Applicant: FUJITSU LIMITED
Inventor： Hiroyasu Yoshikawa , Yukio Hirai
CPC classification number: G06T5/001 , G06T5/002 , G06T5/003 , G06T5/20 , G06T2207/20021
Abstract: An image processing apparatus includes a calculation section configured to calculate filtering coefficients of a filter with a first area in an image that is partitioned into multiple first areas including the first area, the image being partitioned differently into multiple second areas, each one of which being covered by several first areas, to calculate a convoluted image of a second area using the filtering coefficients calculated with the first areas covering a part of the second area, the calculation being executed for the several first areas covering distinct parts of the second area, respectively; and an interpolation section configured to interpolate a pixel in the second area using pixels at the same position in the convoluted images of the second area which are convoluted with the respective filtering coefficients.
Abstract translation: 一种图像处理装置，包括：计算部，其被配置为计算滤波器的滤波系数，该滤波器的图像中的第一区域被划分为包括所述第一区域的多个第一区域，所述图像被不同地分割成多个第二区域，每个第二区域是 由几个第一区域覆盖，使用覆盖第二区域的一部分的第一区域计算的滤波系数来计算第二区域的卷积图像，分别对覆盖第二区域的不同部分的几个第一区域执行计算 ; 以及插值部件，被配置为使用与各个滤波系数进行卷积的第二区域的卷积图像中的相同位置处的像素来内插第二区域中的像素。
-
2.发明授权Gaze detection apparatus, gaze detection computer program, and display apparatus 有权注视检测装置，注视检测计算机程序和显示装置公开(公告)号：US09473696B2
公开(公告)日：2016-10-18
申请号：US13947605
申请日：2013-07-22
Applicant: FUJITSU LIMITED
Inventor： Takuya Kamimura , Hiroyasu Yoshikawa
CPC classification number: H04N5/23219
Abstract: A gaze detection apparatus includes a light source which illuminates a user's eye, an imaging unit which generates an image by capturing the image of a user's face, and a processor adapted to: estimate a distance between the imaging unit and the user's face; calculate the ratio of the distance between the imaging unit and the user's face to a separation distance between the imaging unit and the light source and determine, based on the ratio, whether a bright pupil phenomenon occurs; detect from the captured image a corneal reflection image of the light source and a center of the user's pupil when determining that the bright pupil phenomenon does not occur; and detect the user's gaze direction or gaze position based on a positional relationship between the center of the user's pupil and the corneal reflection image when determining that the bright pupil phenomenon does not occur.
Abstract translation: 注视检测装置包括照亮用户的眼睛的光源，通过拍摄用户脸部的图像来生成图像的成像单元和适于：估计成像单元与使用者脸部之间的距离的处理器; 计算成像单元与用户面部之间的距离与成像单元与光源之间的间隔距离的比例，并根据该比率确定是否发生明亮的瞳孔现象; 当确定不发生明亮的瞳孔现象时，从拍摄图像中检测光源的角膜反射图像和用户瞳孔的中心; 并且当确定不发生明亮的瞳孔现象时，基于用户瞳孔的中心与角膜反射图像之间的位置关系来检测用户的注视方向或注视位置。
-
3.发明申请公开(公告)号：US20130155254A1
公开(公告)日：2013-06-20
申请号：US13664912
申请日：2012-10-31
Applicant: FUJITSU LIMITED
Inventor： Nobuyuki KANTO , Masayoshi Shimizu , Hiroyasu Yoshikawa , Yukio Hirai , Takuya Kamimura
IPC: H04N5/33
Abstract: An imaging unit has sensitivity to visible light and infrared light and captures an image. When the distribution of the color of each pixel in the image captured by the imaging unit is calculated, a deriving unit derives a predetermined feature amount indicating the range of the color distribution. An estimating unit estimates a lighting environment during imaging based on the feature amount derived by the deriving unit.
Abstract translation: 成像单元对可见光和红外光具有敏感度并捕获图像。 当计算由成像单元拍摄的图像中的每个像素的颜色的分布时，导出单元导出指示颜色分布的范围的预定特征量。 估计单元基于由导出单元导出的特征量来估计成像期间的照明环境。
-
公开(公告)号：US20130088613A1
公开(公告)日：2013-04-11
申请号：US13632695
申请日：2012-10-01
Applicant: FUJITSU LIMITED
Inventor： Nobuyuki Kanto , Masayoshi Shimizu , Hiroyasu Yoshikawa , Yukio Hirai , Takuya Kamimura
IPC: H04N9/73
CPC classification number: H04N1/6027 , H04N9/045
Abstract: A determining unit determines which one of a plurality of divided areas in color distribution a value of an input image single falls within. The plurality of divided areas are defined for each of R, G, and B color components, and the determining unit determines, for each color component, which one of the plurality of divided areas the value of the input image signal falls within. An image correcting unit reads correction coefficients corresponding to the determined divided area for each color component from a correction coefficient storing unit in which correction coefficients are registered for each of the color components and each of the divided areas. The image correcting unit calculates each of R, G, and B output values of an output image signal using the correction coefficients of the individual color components read from the correction coefficient storing unit.
Abstract translation: 确定单元确定颜色分布中的多个分割区域中的哪一个是输入图像单个的值在其内的哪一个。 针对R，G和B颜色分量中的每一个分配多个分割区域，并且确定单元针对每个颜色分量确定输入图像信号的值的哪一个分割区域。 图像校正单元从校正系数存储单元中读取对于每个颜色分量所确定的分割区域的校正系数，其中针对每个颜色分量和每个分割区域登记校正系数。 图像校正单元使用从校正系数存储单元读取的各个颜色分量的校正系数来计算输出图像信号的R，G和B输出值。
-
5.发明授权Device and computer-readable recording medium for imaging and image processing with color correction 有权用于具有色彩校正的成像和图像处理的装置和计算机可读记录介质公开(公告)号：US08810694B2
公开(公告)日：2014-08-19
申请号：US13657192
申请日：2012-10-22
Applicant: Fujitsu Limited
Inventor： Yukio Hirai , Masayoshi Shimizu , Hiroyasu Yoshikawa , Nobuyuki Kanto , Takuya Kamimura
CPC classification number: H04N5/357 , G06T5/002 , H04N1/6033 , H04N9/045 , H04N9/67 , H04N17/002
Abstract: A memory unit stores as conversion information which is derived based on a correction coefficient which provides, in a predetermined allowable range from a minimum value, a sum of a norm of a difference between a color difference of a correction target color and a target color which is a target of correction of the correction target color and a correction color obtained by correcting the correction target color using the correction coefficient, and a norm of the correction coefficient to which a predetermined weight coefficient is applied. A color correcting unit converts a color of each pixel of an image captured by an image capturing unit based on the conversion information.
Abstract translation: 存储单元存储作为基于校正系数导出的转换信息，该校正系数在最小值的预定允许范围内提供校正目标色的色差和目标色的差的范数之和， 是校正目标颜色的校正目标和通过使用校正系数校正校正目标颜色而获得的校正颜色，以及应用了预定加权系数的校正系数的范数。 颜色校正单元基于转换信息来转换由图像捕获单元捕获的图像的每个像素的颜色。
-
Patent Agency Ranking
公开(公告)号：US20130343651A1
公开(公告)日：2013-12-26
申请号：US13900875
申请日：2013-05-23
Applicant: FUJITSU LIMITED
Inventor： Yukio Hirai , Hiroyasu Yoshikawa
IPC: G06T5/00
CPC classification number: G06T5/003 , G06T5/20 , G06T2207/20012 , H04N17/002
Abstract: An image processing apparatus includes: a processor coupled to a memory, configured to: perform an analysis of resolution in at least two directions of an image which is taken from a subject having a radial pattern, and determine filter data containing an adjusted weight coefficient which is obtained by adjusting a weight coefficient in one of the two directions, whichever has lower resolution, based on a result of the analysis of the image corrected by filtering on the image in accordance with a blurring function of the image.
Abstract translation: 一种图像处理装置，包括：耦合到存储器的处理器，被配置为：对从具有径向图案的对象取得的图像的至少两个方向执行分辨率的分析，并且确定包含经校正的权重系数的滤波器数据， 基于通过根据图像的模糊功能对图像进行滤波而校正的图像的分析结果，通过调整两个方向中的一个方向上的权重系数（取决于较低分辨率）来获得。
-
公开(公告)号：US20130342709A1
公开(公告)日：2013-12-26
申请号：US13863497
申请日：2013-04-16
Applicant: FUJITSU LIMITED
Inventor： Yukio Hirai , Hiroyasu Yoshikawa
IPC: H04N5/225
CPC classification number: G06T5/003 , G06T2207/10004 , H04N5/225
Abstract: An image processing apparatus includes an acquisition unit configured to acquire a first finite spatial filter having image resolution anisotropy, and a calculation unit configured to compute a second spatial filter by convolving a finite filter with respect to the first spatial filter, the finite filter having a sum of elements being 0 and at least two of the elements being non-0.
Abstract translation: 一种图像处理装置，包括：获取单元，被配置为获取具有图像分辨率各向异性的第一有限空间滤波器;以及计算单元，被配置为通过相对于所述第一空间滤波器卷积有限滤波器来计算第二空间滤波器，所述有限滤波器具有 元素的和为0，并且至少两个元素不为0。
-
8.发明授权Anisotropy in resolution spatial filter image processing apparatus, and information processing method 有权分辨率空间滤波图像处理装置中的各向异性，以及信息处理方法公开(公告)号：US09418406B2
公开(公告)日：2016-08-16
申请号：US14082536
申请日：2013-11-18
Applicant: FUJITSU LIMITED
Inventor： Yukio Hirai , Hiroyasu Yoshikawa
CPC classification number: G06T5/003 , G06T5/20 , G06T2207/20012 , H04N5/217 , H04N5/3572
Abstract: An information processing method includes calculating a second spatial filter having a size of the number of elements larger than a blur size of an image using a finite first spatial filter having an anisotropy in resolution of the image and a finite filter in which a value of a total sum of elements is zero and at least two elements have a non-zero value, and generating a plurality of spatial filters having a predetermined number of elements or less from the second spatial filter.
Abstract translation: 信息处理方法包括使用具有图像分辨率的各向异性的有限第一空间滤波器以及有限滤波器来计算具有大于图像的模糊大小的元素的数量的尺寸的第二空间滤波器，其中值 元素的总和为零，并且至少两个元素具有非零值，并且从第二空间滤波器生成具有预定数量的元素或更小的多个空间滤波器。
-
公开(公告)号：US09202264B2
公开(公告)日：2015-12-01
申请号：US13900875
申请日：2013-05-23
Applicant: FUJITSU LIMITED
Inventor： Yukio Hirai , Hiroyasu Yoshikawa
CPC classification number: G06T5/003 , G06T5/20 , G06T2207/20012 , H04N17/002
Abstract: An image processing apparatus includes: a processor coupled to a memory, configured to: perform an analysis of resolution in at least two directions of an image which is taken from a subject having a radial pattern, and determine filter data containing an adjusted weight coefficient which is obtained by adjusting a weight coefficient in one of the two directions, whichever has lower resolution, based on a result of the analysis of the image corrected by filtering on the image in accordance with a blurring function of the image.
Abstract translation: 一种图像处理装置包括：耦合到存储器的处理器，被配置为：对从具有径向图案的对象取得的图像的至少两个方向执行分辨率的分析，并且确定包含经校正的权重系数的滤波器数据， 基于通过根据图像的模糊功能对图像进行滤波而校正的图像的分析结果，通过调整两个方向中的一个方向上的权重系数（取决于较低分辨率）来获得。
-
公开(公告)号：US09042642B2
公开(公告)日：2015-05-26
申请号：US13632695
申请日：2012-10-01
Applicant: FUJITSU LIMITED
Inventor： Nobuyuki Kanto , Masayoshi Shimizu , Hiroyasu Yoshikawa , Yukio Hirai , Takuya Kamimura
CPC classification number: H04N1/6027 , H04N9/045
Abstract: A determining unit determines which one of a plurality of divided areas in color distribution a value of an input image single falls within. The plurality of divided areas are defined for each of R, G, and B color components, and the determining unit determines, for each color component, which one of the plurality of divided areas the value of the input image signal falls within. An image correcting unit reads correction coefficients corresponding to the determined divided area for each color component from a correction coefficient storing unit in which correction coefficients are registered for each of the color components and each of the divided areas. The image correcting unit calculates each of R, G, and B output values of an output image signal using the correction coefficients of the individual color components read from the correction coefficient storing unit.
Abstract translation: 确定单元确定颜色分布中的多个分割区域中的哪一个是输入图像单个的值在其内的哪一个。 针对R，G和B颜色分量中的每一个分配多个分割区域，并且确定单元针对每个颜色分量确定输入图像信号的值的哪一个分割区域。 图像校正单元从校正系数存储单元中读取对于每个颜色分量所确定的分割区域的校正系数，其中针对每个颜色分量和每个分割区域登记校正系数。 图像校正单元使用从校正系数存储单元读取的各个颜色分量的校正系数来计算输出图像信号的R，G和B输出值。
-
-
-
-
-
-
-
-
-
Search Results
16
records
(took 0.016 seconds)
