公开(公告)号：US09495578B2

公开(公告)日：2016-11-15

申请号：US13344191

申请日：2012-01-05

申请人： Miyuki Okada ,  Ken Nakajima

发明人： Miyuki Okada ,  Ken Nakajima

IPC分类号： G06K9/00 ,  H04N5/232

CPC分类号： G06K9/00228 ,  H04N5/23219

摘要： A face detection device for detecting the face of a person in an input image may include the following elements: a face detection circuit including a hardware circuit configured to detect a face in an input image; a signal processing circuit configured to perform signal processing based on an input image signal in accordance with a rewritable program including a face detection program for detecting a face in an input image; and a controller configured to allow the face detection circuit and the signal processing circuit to perform face detection on an image of a frame or on respective images of adjacent frames among consecutive frames, and to control face detection by the signal processing circuit on the basis of a face detection result obtained by the face detection circuit.

摘要翻译： 用于检测输入图像中的人脸的脸部检测装置可以包括以下元件：面部检测电路，包括被配置为检测输入图像中的脸部的硬件电路; 信号处理电路，被配置为根据包括用于检测输入图像中的面部的面部检测程序的可重写程序，基于输入图像信号执行信号处理; 以及控制器，其被配置为允许所述面部检测电路和所述信号处理电路对连续帧中的帧的图像或相邻帧的各个图像执行面部检测，并且基于所述信号处理电路来控制所述信号处理电路的面部检测 由脸部检测电路获得的面部检测结果。

公开(公告)号：US08295595B2

公开(公告)日：2012-10-23

申请号：US12279190

申请日：2007-09-20

申请人： Nobuyuki Matsushita ,  Ken Nakajima ,  Mikiko Yamada

发明人： Nobuyuki Matsushita ,  Ken Nakajima ,  Mikiko Yamada

IPC分类号： G06K9/00 ,  G06K9/40 ,  H04N9/64

CPC分类号： H04N9/045 ,  H04N5/217 ,  H04N2101/00 ,  H04N2209/046

摘要： A method and apparatus to remove color noise included in raw data while effectively preventing image quality degradation. For interest pixels serially set onto a mosaic image formed of raw data, conversion is executed into a pixel value for noise removal based on a processing reference pixel value having a unified color signal component in each interest pixel, noise is removed from the pixel value for noise removal, and the pixel value for noise removal with noise removed is converted into the source pixel value, whereby only color noise can be removed without affecting a luminance signal.

摘要翻译： 一种消除原始数据中包含的颜色噪声同时有效防止图像质量下降的方法和装置。 对于由原始数据构成的马赛克图像串行设置的兴趣像素，基于在每个兴趣像素中具有统一颜色信号分量的处理参考像素值，将转换执行为用于噪声去除的像素值，从像素值中去除噪声 噪声消除，并且将去除噪声的噪声去除的像素值转换为源像素值，由此仅可以去除颜色噪声而不影响亮度信号。

公开(公告)号：US08102434B2

公开(公告)日：2012-01-24

申请号：US11556304

申请日：2006-11-03

申请人： Tsuyoshi Yoshida ,  Nobuyuki Sato ,  Ken Nakajima

发明人： Tsuyoshi Yoshida ,  Nobuyuki Sato ,  Ken Nakajima

IPC分类号： H04N5/225 ,  H04N5/00

CPC分类号： H04N5/232 ,  H04N5/23245 ,  H04N9/045 ,  H04N9/07

摘要： An imaging apparatus capturing and recording a moving image and a static image at the same time includes a converter converting an optical image of a subject into a pixel signal, performing decimation on the pixel signal and outputting the pixel signal that has been subjected to decimation when recording of the moving image is performed but recording of the static image is not performed, and outputting the pixel signal that has not been subjected to decimation when recording of the moving image and the static image is performed; an eliminating section performing decimation on the pixel signal output from the converter only when recording of the moving image and the static image is performed; a moving image data generator generating moving image data; a static image data generator generating static image data; and a recorder recording the moving image data and the static image data.

摘要翻译： 同时拍摄和记录运动图像和静态图像的成像装置包括将被摄体的光学图像转换为像素信号的转换器，对像素信号执行抽取并输出已经进行抽取的像素信号，当 执行运动图像的记录，而不执行静态图像的记录，并且当执行运动图像和静态图像的记录时，输出未进行抽取的像素信号; 消除部，仅在进行运动图像和静态图像的记录时，对从转换器输出的像素信号进行抽取; 生成运动图像数据的运动图像数据生成器; 产生静态图像数据的静态图像数据发生器; 以及记录运动图像数据和静态图像数据的记录器。

公开(公告)号：US20090067744A1

公开(公告)日：2009-03-12

申请号：US12264500

申请日：2008-11-04

申请人： Isao KAWANISHI ,  Miyuki Okada ,  Nobuyuki Sato ,  Ken Nakajima

发明人： Isao KAWANISHI ,  Miyuki Okada ,  Nobuyuki Sato ,  Ken Nakajima

IPC分类号： G06K9/40

CPC分类号： H04N5/243 ,  H04N5/217 ,  H04N5/23296 ,  H04N9/646

摘要： For each pixel read from an imaging device, a horizontal counter value and a vertical counter value corresponding to the pixel are supplied from a signal generator to a distance computation section via an optical-axis-center coordinate setting section and an up-and-down and right-and-left weighting section. In the distance computation section, the distance to the optical-axis center is computed, and correction coefficients for the zoom wide end and for the zoom tele end, which correspond to the distance, are obtained by look-up tables. The two obtained correction coefficients are blended at a mixture ratio determined by a blend ratio setting section. The blended shading correction coefficients are gain adjusted by a gain adjustment section, after which they are supplied to a correction section. As a result, a correction corresponding to the distance to the optical-axis-center position is performed on the signal of each pixel supplied from an imaging section.

摘要翻译： 对于从成像装置读取的每个像素，将与像素相对应的水平计数器值和垂直计数器值经由光轴中心坐标设定部和上下方向从信号发生器供给到距离计算部 和右，右加权部分。 在距离计算部分中，计算到光轴中心的距离，并且通过查找表获得与距离对应的变焦宽端和变焦远端的校正系数。 将两个获得的校正系数以由混合比设置部确定的混合比混合。 混合阴影校正系数是通过增益调整部分调整的增益，之后将它们提供给校正部分。 结果，对从摄像部提供的每个像素的信号执行与光轴中心位置的距离相对应的校正。

公开(公告)号：US07433531B2

公开(公告)日：2008-10-07

申请号：US10468981

申请日：2002-12-10

申请人： Isao Kawanishi ,  Tomohiko Naruse ,  Ken Nakajima

发明人： Isao Kawanishi ,  Tomohiko Naruse ,  Ken Nakajima

IPC分类号： G06K9/36 ,  G06K9/38 ,  G06K9/46 ,  H04N1/41

CPC分类号： H04N5/772 ,  H04N5/907 ,  H04N9/8047 ,  H04N9/8227 ,  H04N19/115 ,  H04N19/117 ,  H04N19/124 ,  H04N19/136 ,  H04N19/149 ,  H04N19/15 ,  H04N19/162 ,  H04N19/176 ,  H04N19/59 ,  H04N19/60 ,  H04N19/80 ,  H04N2101/00 ,  H04N2201/33357

摘要： The present invention relates to an image capturing apparatus and method in which the image recording time is reduced and the memory capacity required for compression is also reduced. A number-of-bytes calculation unit 302 determines the number of bytes after compression based on an integrated value of high-frequency integrated data supplied from a high-frequency integration processor. Based on the determined number of bytes, a Q-scale calculation unit 303 determines a Q-scale based on which the image data can be compressed one time to a predetermined data size. A Q-table generation unit 304 generates a Q-table based on the Q-scale. A DCT unit 321 performs a discrete cosine transform on the input image data. A quantization processor 322 adjusts the compression ratio of the image data based on the up-to-date Q-table supplied from the Q-table generation unit 304. A variable-length coding processor 323 encodes the image data with variable length coding such as Huffman coding, and outputs the resulting compressed image data. The present invention is applicable to digital cameras.

摘要翻译： 本发明涉及图像记录时间缩短并且压缩所需的存储容量也降低的图像捕获装置和方法。 字节数计算单元302基于从高频积分处理器提供的高频综合数据的积分值来确定压缩后的字节数。 基于所确定的字节数，Q尺度计算单元303根据其可以将图像数据压缩一次到预定数据大小来确定Q标度。 Q表生成单元304基于Q量表生成Q表。 DCT单元321对输入图像数据执行离散余弦变换。 量化处理器322基于从Q表生成单元304提供的最新的Q表来调整图像数据的压缩比。 可变长度编码处理器323对诸如霍夫曼编码之类的可变长度编码的图像数据进行编码，并输出所得到的压缩图像数据。 本发明适用于数码相机。

公开(公告)号：US20070223824A1

公开(公告)日：2007-09-27

申请号：US11710622

申请日：2007-02-23

申请人： Fumiaki Kato ,  Ken Nakajima

发明人： Fumiaki Kato ,  Ken Nakajima

IPC分类号： G06K9/36

CPC分类号： H04N19/98 ,  H04N5/335 ,  H04N5/772 ,  H04N5/775 ,  H04N5/85 ,  H04N5/907 ,  H04N9/7921 ,  H04N9/8047 ,  H04N2101/00 ,  H04N2201/0084

摘要： An image capturing apparatus for capturing an image with solid state imaging devices may include a compressing section, a memory, a decompressing section, and a signal processing section. The compressing section may compress digitalized data of an image captured with the solid state imaging devices. The memory may temporarily store the compressed image data that is compressed by the compressing section. The decompressing section may decompress the compressed image data that is read out from the memory. The signal processing section may perform an image quality correction operation on the image data decompressed by the decompressing section. The compressed image data may contain a maximum value and a minimum value of pixel data in a block, information regarding positions of the maximum value and the minimum value in the block, and quantized data.

摘要翻译： 用于使用固态成像装置捕获图像的图像捕获装置可以包括压缩部分，存储器，解压缩部分和信号处理部分。 压缩部分可以压缩用固态成像装置捕获的图像的数字化数据。 存储器可以临时存储由压缩部分压缩的压缩图像数据。 解压缩部分可以解压缩从存储器读出的压缩图像数据。 信号处理部分可以对由解压缩部分解压缩的图像数据执行图像质量校正操作。 压缩图像数据可以包含块中的像素数据的最大值和最小值，关于块中的最大值和最小值的位置的信息以及量化数据。

公开(公告)号：US20060082665A1

公开(公告)日：2006-04-20

申请号：US10540404

申请日：2003-12-02

申请人： Takami Mizukura ,  Naoya Katoh ,  Ken Nakajima ,  Hiroki Nagahama

发明人： Takami Mizukura ,  Naoya Katoh ,  Ken Nakajima ,  Hiroki Nagahama

IPC分类号： H04N9/04

CPC分类号： H04N9/045

摘要： In an image pick-up apparatus including an image pick-up unit comprised of color filters having different spectral characteristics, and serving to pick up image of object, there are provided an adjustment unit for adjusting color reproduction value and noise value representing noise feeling, a matrix coefficient determination unit for determining matrix coefficients on the basis of adjustment of the adjustment unit, and a matrix transform processing unit for performing matrix transform processing with respect to image which has been picked up at the image pick-up device unit on the basis of the matrix coefficients.

摘要翻译： 在包括由具有不同光谱特性的滤色器构成的并用于拾取物体图像的图像拾取单元的图像拾取装置中，提供了一种用于调节颜色再现值和表示噪声感觉的噪声值的调整单元， 矩阵系数确定单元，用于基于调整单元的调整来确定矩阵系数;以及矩阵变换处理单元，用于基于在图像拾取装置单元拾取的图像执行矩阵变换处理 的矩阵系数。

公开(公告)号：US20050089241A1

公开(公告)日：2005-04-28

申请号：US10921284

申请日：2004-08-19

申请人： Isao Kawanishi ,  Miyuki Okada ,  Nobuyuki Sato ,  Ken Nakajima

发明人： Isao Kawanishi ,  Miyuki Okada ,  Nobuyuki Sato ,  Ken Nakajima

IPC分类号： G06T1/00 ,  G06T5/00 ,  H04N1/19 ,  H04N1/40 ,  H04N1/409 ,  H04N5/225 ,  H04N5/243 ,  H04N9/07 ,  H04N9/64 ,  H04N101/00

CPC分类号： H04N5/243 ,  H04N5/217 ,  H04N5/23296 ,  H04N9/646

摘要： For each pixel read from an imaging device, a horizontal counter value and a vertical counter value corresponding to the pixel are supplied from a signal generator to a distance computation section via an optical-axis-center coordinate setting section and an up-and-down and right-and-left weighting section. In the distance computation section, the distance to the optical-axis center is computed, and correction coefficients for the zoom wide end and for the zoom tele end, which correspond to the distance, are obtained by look-up tables. The two obtained correction coefficients are blended at a mixture ratio determined by a blend ratio setting section. The blended shading correction coefficients are gain adjusted by a gain adjustment section, after which they are supplied to a correction section. As a result, a correction corresponding to the distance to the optical-axis-center position is performed on the signal of each pixel supplied from an imaging section.

摘要翻译： 对于从成像装置读取的每个像素，将与像素相对应的水平计数器值和垂直计数器值经由光轴中心坐标设定部和上下方向从信号发生器供给到距离计算部 和右，右加权部分。 在距离计算部分中，计算到光轴中心的距离，并且通过查找表获得与距离对应的变焦宽端和变焦远端的校正系数。 将两个获得的校正系数以由混合比设置部确定的混合比混合。 混合阴影校正系数是通过增益调整部分调整的增益，之后将它们提供给校正部分。 结果，对从摄像部提供的每个像素的信号执行与光轴中心位置的距离相对应的校正。

公开(公告)号：US06243487B1

公开(公告)日：2001-06-05

申请号：US09050355

申请日：1998-03-31

申请人： Ken Nakajima

发明人： Ken Nakajima

IPC分类号： G03F720

CPC分类号： B82Y10/00 ,  B82Y40/00 ,  H01J37/3174 ,  H01J2237/31771 ,  Y10S430/133 ,  Y10S430/143

摘要： Disclosed is a pattern exposure method for conducting an overlay exposure to a base pattern previously formed by using electron beam, which has the steps of: 1) dividing a group of patterns formed on the base pattern into arbitrary regions; 2) calculating an area density of the base pattern in each of the arbitrary regions; 3) generating a pattern according to the area density; 4) conducting sub-exposure using the pattern; and 5) conducting main exposure to expose a desired pattern.

摘要翻译： 公开了一种用于对通过使用电子束预先形成的基底图案进行覆盖曝光的图案曝光方法，其具有以下步骤：1）将形成在基底图案上的一组图案划分成任意区域; 2）计算每个任意区域中的基本图案的面积密度; 3）根据面积密度生成图案; 4）使用图案进行次曝光; 和5）进行主要曝光以暴露期望的图案。

公开(公告)号：US6007949A

公开(公告)日：1999-12-28

申请号：US924535

申请日：1997-09-05

申请人： Ken Nakajima

发明人： Ken Nakajima

IPC分类号： G03F9/00 ,  G03F1/20 ,  H01L21/027 ,  H01L21/8242 ,  H01L27/108

CPC分类号： G03F1/20 ,  H01J2237/0453 ,  H01J2237/31794

摘要： In a method of forming a mask from a design pattern, a mask pattern is repeated in the design pattern and is checked about whether or not the mask pattern has the same line width at a border. If the mask pattern having the same line width is divided, a portion having the same line width adjacent to the border is removed from the mask pattern. The removed mask pattern is actually used as the mask for the electron beam exposure.As a result, when electron beam exposure is made by the use of the mask, exposed patterns are partially overlapped with each other at adjacent ones of the exposed patterns and contiguous to each other at portions having different line widths. With this structure, the electron beam exposure is performed without registration errors between the exposed patterns contiguous to each other.

摘要翻译： 在从设计图案形成掩模的方法中，在设计图案中重复掩模图案，并且检查掩模图案是否在边界处具有相同的线宽。 如果具有相同线宽的掩模图案被分割，则与掩模图案相邻的具有与边界相邻的线宽的部分被去除。 去除的掩模图案实际上用作电子束曝光的掩模。 结果，当通过使用掩模进行电子束曝光时，暴露图案在相邻的曝光图案处彼此部分地重叠，并且在具有不同线宽的部分处彼此邻接。 利用这种结构，在彼此相邻的曝光图案之间执行电子束曝光而没有配准误差。