公开(公告)号：US07180629B1

公开(公告)日：2007-02-20

申请号：US09719950

申请日：2000-04-28

申请人： Kenichi Nishio ,  Eiji Mizutani

发明人： Kenichi Nishio ,  Eiji Mizutani

IPC分类号： H04N1/48 ,  H04N1/60 ,  H04N5/235 ,  H04N9/73 ,  G06T1/40

CPC分类号： H04N9/735

摘要： Methods and apparatus for color correction of color image data obtained by an electronic camera determine a correction to data representative of the color image based upon an estimated illuminant using a neural network, multilayer perceptron models and/or coactive neuro-fuzzy inference system models, and apply the correction to the data representative of the color image. Data representative of the color corrected data may be recorded or transmitted. A method of recording image data obtained by an electronic camera captures a color image, generates data representative of the image, estimates an illuminant for the captured color image, generates data representative of the estimated illuminant and records the data representative of the image with the data representative of the estimated illuminant. A method of transmitting image data obtained by an electronic camera captures a color image, generates data representative of the image, estimates an illuminant for the captured color image, generates data representative of the estimated illuminant and transmits the data representative of the image with the data representative of the estimated illuminant.

摘要翻译： 通过电子照相机获得的彩色图像数据的颜色校正方法和装置基于使用神经网络，多层感知器模型和/或共同作用的神经 - 模糊推理系统模型的估计光源来确定代表彩色图像的数据的校正，以及 将该校正应用于代表彩色图像的数据。 可以记录或发送表示颜色校正数据的数据。 记录由电子照相机获取的图像数据的方法捕获彩色图像，生成表示图像的数据，估计拍摄的彩色图像的光源，生成表示估计的光源的数据，并将表示图像的数据记录在数据中 代表估计的光源。 发送由电子照相机获取的图像数据的方法捕获彩色图像，产生表示图像的数据，估计拍摄的彩色图像的光源，产生表示估计的光源的数据，并将表示图像的数据发送给数据 代表估计的光源。

公开(公告)号：US08731281B2

公开(公告)日：2014-05-20

申请号：US13065729

申请日：2011-03-29

申请人： Xiaogang Dong ,  Farhan Baqai ,  Kenichi Nishio

发明人： Xiaogang Dong ,  Farhan Baqai ,  Kenichi Nishio

IPC分类号： G06K9/00

CPC分类号： G06T3/4015 ,  G06T5/002 ,  G06T2207/20064 ,  H04N9/045

摘要： A system and method for effectively performing wavelet transforms on incomplete image data includes an image processor that performs a green-pixel transformation procedure on incomplete color pixel matrices. The image processor then rearranges red, blue and transformed green-pixel into four quadrants of contiguous pixels and applies some two dimensional (2D) wavelet thresholding schemes on each quadrant. After thresholding, an inverse procedure is applied to reconstruct the pixel values on the incomplete color pixel matrices. For further de-correlation of image data, the image processor may stack similar image patches in a three dimensional (3D) array and apply incomplete-data wavelet thresholding on the 3D array. The incomplete-data wavelet thresholding procedure may be put in an improved local similarity measurement framework to achieve better performance of image processing tasks. A CPU device typically controls the image processor to effectively perform the image processing procedure.

摘要翻译： 用于在不完整图像数据上有效地执行小波变换的系统和方法包括对不完整的彩色像素矩阵执行绿色像素变换过程的图像处理器。 然后，图像处理器将红色，蓝色和变换的绿色像素重新排列成四个相邻像素的象限，并在每个象限上应用一些二维（2D）小波阈值方案。 阈值处理后，应用逆过程来重构不完全色彩像素矩阵上的像素值。 为了图像数据的进一步去相关，图像处理器可以在三维（3D）阵列中堆叠相似的图像块，并在3D阵列上应用不完全数据小波阈值。 不完全数据小波阈值处理可以放在改进的局部相似度测量框架中，以实现更好的图像处理任务性能。 CPU设备通常控制图像处理器以有效地执行图像处理过程。

公开(公告)号：US08730357B2

公开(公告)日：2014-05-20

申请号：US13096438

申请日：2011-04-28

申请人： Kenichi Nishio ,  Hyongmyong Kang ,  Hiroyuki Kinoshita ,  Tsukasa Hashino ,  Takuya Kato ,  Gentaro Irisawa ,  Atsuo Minato

发明人： Kenichi Nishio ,  Hyongmyong Kang ,  Hiroyuki Kinoshita ,  Tsukasa Hashino ,  Takuya Kato ,  Gentaro Irisawa ,  Atsuo Minato

IPC分类号： H04N5/217

CPC分类号： H04N9/045 ,  H04N9/07

摘要： An image processing method includes the steps of calculating, with respect to a processing-target pixel in an input image signal, a concentric aberration correction amount for concentric aberration that is a component of magnification chromatic aberration, the concentric aberration causing a color shift to occur in a concentric manner from the center of an image, calculating, with respect to the processing-target pixel, a uniform aberration correction amount for uniform aberration that is a component of magnification chromatic aberration, the uniform aberration causing a color shift direction and a color shift amount to uniformly occur on a whole image, and correcting a pixel value of the processing-target pixel on the basis of the calculated concentric aberration correction amount and the calculated uniform aberration correction amount.

摘要翻译： 一种图像处理方法包括以下步骤：相对于输入图像信号中的处理目标像素，计算作为倍率色像差的分量的同心像差的同心像差校正量，产生发生色偏的同心像差 以图像中心的同心方式，相对于处理对象像素计算作为倍率色像差的分量的均匀像差的均匀像差校正量，引起色移方向和颜色的均匀像差 基于所计算的同心像差校正量和计算出的均匀像差校正量来校正处理对象像素的像素值。

公开(公告)号：US07782367B2

公开(公告)日：2010-08-24

申请号：US11818796

申请日：2007-06-14

申请人： Alexander Berestov ,  Ted J. Cooper ,  Kenichi Nishio ,  Masaya Kinoshita ,  Ting Zhang

发明人： Alexander Berestov ,  Ted J. Cooper ,  Kenichi Nishio ,  Masaya Kinoshita ,  Ting Zhang

IPC分类号： H04N5/228

CPC分类号： G06K9/209 ,  H04N9/045 ,  H04N9/735 ,  H04N17/002

摘要： A system for and method of determining calibration parameters while only capturing a single image is described herein. Furthermore, traditional calibration algorithms are avoided by the direct determination of the calibration parameters. The determination of the calibration parameters is possible by first determining a training data set from images acquired of a variety of objects with a multitude of colors. Then, using the training data set, regression coefficients are generated. A camera to be calibrated then acquires only one set of image information such as a single picture. Then, using the regression coefficients and the acquired information, the calibration parameters are directly estimated for that camera.

摘要翻译： 本文描述了仅捕获单个图像的确定校准参数的系统和方法。 此外，通过直接确定校准参数来避免传统的校准算法。 通过首先从具有多种颜色的各种物体获得的图像中确定训练数据集，可以确定校准参数。 然后，使用训练数据集，生成回归系数。 要校准的相机只能获取一组图像信息，例如单张图像。 然后，使用回归系数和所获取的信息，直接针对该相机估计校准参数。

公开(公告)号：US20080310710A1

公开(公告)日：2008-12-18

申请号：US11818796

申请日：2007-06-14

申请人： Alexander Berestov ,  Ted J. Cooper ,  Kenichi Nishio ,  Masaya Kinoshita ,  Ting Zhang

发明人： Alexander Berestov ,  Ted J. Cooper ,  Kenichi Nishio ,  Masaya Kinoshita ,  Ting Zhang

IPC分类号： G06K9/00

CPC分类号： G06K9/209 ,  H04N9/045 ,  H04N9/735 ,  H04N17/002

摘要： A system for and method of determining calibration parameters while only capturing a single image is described herein. Furthermore, traditional calibration algorithms are avoided by the direct determination of the calibration parameters. The determination of the calibration parameters is possible by first determining a training data set from images acquired of a variety of objects with a multitude of colors. Then, using the training data set, regression coefficients are generated. A camera to be calibrated then acquires only one set of image information such as a single picture. Then, using the regression coefficients and the acquired information, the calibration parameters are directly estimated for that camera.

摘要翻译： 本文描述了仅捕获单个图像的确定校准参数的系统和方法。 此外，通过直接确定校准参数来避免传统的校准算法。 通过首先从具有多种颜色的各种物体获得的图像中确定训练数据集，可以确定校准参数。 然后，使用训练数据集，生成回归系数。 要校准的相机只能获取一组图像信息，例如单张图像。 然后，使用回归系数和所获取的信息，直接针对该相机估计校准参数。

公开(公告)号：US08446477B2

公开(公告)日：2013-05-21

申请号：US13190900

申请日：2011-07-26

申请人： Gentaro Irisawa ,  Atsuo Kikuchi ,  Takuya Kato ,  Atsuo Minato ,  Kenichi Nishio ,  Hiroyuki Kinoshita

发明人： Gentaro Irisawa ,  Atsuo Kikuchi ,  Takuya Kato ,  Atsuo Minato ,  Kenichi Nishio ,  Hiroyuki Kinoshita

IPC分类号： H04N5/228 ,  H04N9/64

CPC分类号： H04N5/23248 ,  H04N5/23258 ,  H04N5/23287 ,  H04N5/3572 ,  H04N9/045

摘要： An imaging apparatus includes: a shift lens driven to be shifted; a uniform aberration correction data storing unit storing uniform aberration correction data for correcting uniform aberration; a shift amount calculating unit calculating an amount of aberration correction shift of the shift lens for correcting uniform aberration using the uniform aberration correction data; a shift lens driving unit driving the shift lens to be shifted according to an amount of driving shift based on the amount of aberration correction shift; a concentric aberration correction amount calculating unit calculating an amount of concentric aberration correction for correcting concentric aberration with respect to a process target pixel of an input image signal; and a pixel value correcting unit correcting a pixel value of the process target pixel on the basis of the calculated amount of concentric aberration correction.

摘要翻译： 一种成像装置，包括：驱动移位的移位透镜; 均匀像差校正数据存储单元，其存储用于校正均匀像差的均匀像差校正数据; 偏移量计算单元，使用所述均匀像差校正数据来计算用于校正均匀像差的移位透镜的像差校正偏移量; 移动透镜驱动单元，根据基于像差校正偏移量的驱动移位量，驱动移位透镜进行移位; 同心像差校正量计算单元，计算相对于输入图像信号的处理目标像素校正同心像差的同心像差校正量; 以及像素值校正单元，其基于计算出的同心像差校正量校正处理目标像素的像素值。

公开(公告)号：US08406548B2

公开(公告)日：2013-03-26

申请号：US13036656

申请日：2011-02-28

申请人： Gazi Ali ,  Kenichi Nishio ,  Florian Ciurea ,  Hideki Nabesako

发明人： Gazi Ali ,  Kenichi Nishio ,  Florian Ciurea ,  Hideki Nabesako

IPC分类号： G06K9/00

CPC分类号： G06T5/002

摘要： A method and apparatus for performing a blur rendering process on an image is disclosed. In one embodiment, the method of performing a blur rendering process includes accessing a filtered image and depth map information, determining a plurality of blending coefficients for computing a weighted sum for the image and filtered image, wherein the plurality of blending coefficients define a substantially smooth transition from at least one first depth class to at least one second depth class and a substantially sharp transition from the at least one second depth class and the at least one first depth class, wherein the at least one first depth class and the at least one second depth class form at least a portion of a plurality of depth classes and combining the image and the filtered image into a resulting image using the plurality of coefficients.

摘要翻译： 公开了一种用于对图像执行模糊渲染处理的方法和装置。 在一个实施例中，执行模糊渲染过程的方法包括访问滤波图像和深度图信息，确定用于计算图像和滤波图像的加权和的多个混合系数，其中多个混合系数限定基本平滑 从至少一个第一深度类别转换到至少一个第二深度类别以及从所述至少一个第二深度类别和所述至少一个第一深度类别的基本上尖锐的转变，其中所述至少一个第一深度类别和所述至少一个 第二深度类别形成多个深度类别的至少一部分，并且使用所述多个系数将所述图像和所述滤波图像合成为所得图像。

公开(公告)号：US20120250976A1

公开(公告)日：2012-10-04

申请号：US13065729

申请日：2011-03-29

申请人： Xiaogang Dong ,  Farhan Baqai ,  Kenichi Nishio

发明人： Xiaogang Dong ,  Farhan Baqai ,  Kenichi Nishio

IPC分类号： G06K9/00

CPC分类号： G06T3/4015 ,  G06T5/002 ,  G06T2207/20064 ,  H04N9/045

摘要： A system and method for effectively performing wavelet transforms on incomplete image data includes an image processor that performs a green-pixel transformation procedure on incomplete color pixel matrices. The image processor then rearranges red, blue and transformed green-pixel into four quadrants of contiguous pixels and applies some two dimensional (2D) wavelet thresholding schemes on each quadrant. After thresholding, an inverse procedure is applied to reconstruct the pixel values on the incomplete color pixel matrices. For further de-correlation of image data, the image processor may stack similar image patches in a three dimensional (3D) array and apply incomplete-data wavelet thresholding on the 3D array. The incomplete-data wavelet thresholding procedure may be put in an improved local similarity measurement framework to achieve better performance of image processing tasks. A CPU device typically controls the image processor to effectively perform the image processing procedure.

摘要翻译： 用于在不完整图像数据上有效地执行小波变换的系统和方法包括对不完整的彩色像素矩阵执行绿色像素变换过程的图像处理器。 然后，图像处理器将红色，蓝色和变换的绿色像素重新排列成四个相邻像素的象限，并在每个象限上应用一些二维（2D）小波阈值方案。 阈值处理后，应用逆过程来重构不完全色彩像素矩阵上的像素值。 为了图像数据的进一步去相关，图像处理器可以在三维（3D）阵列中堆叠相似的图像块，并在3D阵列上应用不完全数据小波阈值。 不完全数据小波阈值处理可以放在改进的局部相似度测量框架中，以实现更好的图像处理任务性能。 CPU设备通常控制图像处理器以有效地执行图像处理过程。

公开(公告)号：US06801254B1

公开(公告)日：2004-10-05

申请号：US09657900

申请日：2000-09-08

申请人： Kenichi Nishio

发明人： Kenichi Nishio

IPC分类号： H04N5217

CPC分类号： H04N5/378 ,  H04N5/361 ,  H04N5/3655

摘要： A GCA+ADC chip 23 on a main circuit board 20 is supplied with an image signal from a CCD circuit board 1 via an input terminal 21 connected to a line 16. Concurrently, the GCA+ADC chip 23 is supplied with a specified DC reference signal REF 13 from the CCD circuit board 1 via an input terminal 22 connected to the other line 17. An internal operation subtracts the DC reference signal REF from the image signal. Drivers 9 and 10 are provided with the same characteristics. If an interfering noise is picked up during transmission, these signals carry the same contents on respective transmission. This interfering noise is canceled during the subtraction inside the GCA+ADC chip 23.

摘要翻译： 在主电路板20上的GCA + ADC芯片23经由连接到线16的输入端21从CCD电路板1提供图像信号。同时，向GCA + ADC芯片23提供指定的直流基准 通过连接到另一条线路17的输入端子22从CCD电路板1发送信号REF 13。内部操作从图像信号中减去DC参考信号REF。 驱动器9和10具有相同的特性。 如果在传输过程中接收到干扰噪声，这些信号在相应的传输中携带相同的内容。 在GCA + ADC芯片23内的减法期间，该干扰噪声被消除。

公开(公告)号：US07847822B2

公开(公告)日：2010-12-07

申请号：US11818801

申请日：2007-06-14

申请人： Alexander Berestov ,  Kenichi Nishio ,  Ted J. Cooper ,  Masaya Kinoshita ,  Ting Zhang

发明人： Alexander Berestov ,  Kenichi Nishio ,  Ted J. Cooper ,  Masaya Kinoshita ,  Ting Zhang

IPC分类号： H04N17/00 ,  H04N17/02

CPC分类号： H04N17/02 ,  H04N9/045 ,  H04N17/002

摘要： A system for and method of calibrating an imaging device efficiently is described herein. The imaging device acquires an image of an object that is more than one color. The information acquired is then transferred to a computing device. The information is then used to generate a set of data which represents information which was not acquired in the image. The set of data is generated based on statistical prediction using a training data set. Using acquired image information and the set of data, an imaging device is able to be calibrated. Since the process of calibration utilizing this method only requires one image to be acquired and a reduced set of image information to be sent to the computing device, the process is more efficient than previous implementations.

摘要翻译： 这里描述了一种有效地校准成像装置的系统和方法。 成像装置获取多于一种颜色的对象的图像。 然后将所获取的信息传送到计算设备。 然后，该信息用于生成表示在图像中未获取的信息的一组数据。 基于使用训练数据集的统计预测生成数据集。 使用获取的图像信息和数据集，能够对成像装置进行校准。 由于使用该方法的校准过程仅需要获取一个图像并且将减少的图像信息集合发送到计算设备，所以该处理比先前的实现更有效。