公开(公告)号：US08144213B2

公开(公告)日：2012-03-27

申请号：US12663948

申请日：2008-10-23

申请人： Ryuji Fuchikami ,  Ikuo Fuchigami ,  Tomoo Kimura

发明人： Ryuji Fuchikami ,  Ikuo Fuchigami ,  Tomoo Kimura

IPC分类号： H04N9/73

CPC分类号： H04N5/2357 ,  H04N19/105 ,  H04N19/172 ,  H04N19/436 ,  H04N19/51 ,  H04N19/573 ,  H04N19/61 ,  H04N19/85

摘要： A frame brightness detecting unit 15 detects a frame brightness value of each of a plurality of image frames. A flicker spectrum detecting unit 16 detects, from the frame brightness values of 512 frames, spectrum values at frequencies, such as 100 Hz, 200 Hz, and 300 Hz. A brightness estimating unit 17 estimates a brightens value of a light source from the spectrum values. An encoding unit 18 uses a reciprocal of the estimated brightens value to determine a reference frame. The encoding unit 18 also uses the reciprocal in an evaluation function for motion vector estimation. The image frame is encoded using the motion vector.

摘要翻译： 帧亮度检测单元15检测多个图像帧中的每一个的帧亮度值。 闪烁频谱检测单元16从512帧的帧亮度值检测频率如100Hz，200Hz和300Hz的频谱值。 亮度估计单元17从光谱值估计光源的增亮值。 编码单元18使用估计亮度值的倒数来确定参考帧。 编码单元18还在用于运动矢量估计的评估函数中使用倒数。 使用运动矢量对图像帧进行编码。

公开(公告)号：US08300692B2

公开(公告)日：2012-10-30

申请号：US12043313

申请日：2008-03-06

申请人： Tsuyoshi Nakamura ,  Ikuo Fuchigami ,  Masao Hamada ,  Tadanori Tezuka

发明人： Tsuyoshi Nakamura ,  Ikuo Fuchigami ,  Masao Hamada ,  Tadanori Tezuka

IPC分类号： H04N11/02

CPC分类号： H04N19/51 ,  G11B27/005

摘要： A moving picture coding method, a moving picture decoding method, a moving picture coding device, and a moving picture decoding device perform skipped display even in a moving picture decoding device having insufficient processing capability and while suppressing an increase in an amount of coded data. A plurality of combined pictures are generated each of which is generated by combining a plurality of pictures into a combined picture having the plurality of pictures as regions. The plurality of the combined pictures is coded and a slice boundary between the regions is set in the combined picture. For each of the plurality of combined pictures, when the combined picture has one or more to-be-displayed region except at least one not-to-be-displayed region, the to-be-displayed region is coded with reference to only a to-be-displayed region of another combined picture. The one or more to-be-displayed region is displayed for skipped reproduction.

摘要翻译： 运动图像编码方法，运动图像解码方法，运动图像编码装置和运动图像解码装置即使在处理能力不足的运动图像解码装置中也能够进行跳过显示，同时抑制编码数据量的增加。 产生多个组合图像，每个组合图像通过将多个图像组合成具有多个图像的组合图像作为区域而产生。 多个组合图像被编码，并且在组合图像中设置区域之间的切片边界。 对于多个组合图像中的每一个，当组合图像具有除了至少一个不被显示区域之外的一个或多个待显示区域时，将仅参考待显示区域 另一组合图片的待显示区域。 显示一个或多个待显示区域以进行跳过再现。

公开(公告)号：US20110205394A1

公开(公告)日：2011-08-25

申请号：US13126561

申请日：2009-11-18

申请人： Ikuo Fuchigami

发明人： Ikuo Fuchigami

IPC分类号： H04N9/73

CPC分类号： H04N5/243 ,  H04N5/2357

摘要： Line integration unit 31 performs integration on digital image signals of a horizontal line. Sampling unit 32 samples, as a second line integral, a first line integral relating to each of at least one but not all of horizontal lines constituting one frame. Memory 33 stores acquired second line integrals corresponding to a plurality of frames. Inter-screen DFT unit 34 performs discrete Fourier transform on a second line integral sequence composed of a second line integral most recently sampled by the sampling unit 32b and the second line integrals stored in the memory 33. Flicker extraction unit 35 extracts information of flicker component based on a result of the discrete Fourier transform. Flicker coefficient calculation unit 36 calculates a flicker coefficient based on the information. Correction operation unit 37 performs operation of reducing the flicker component on the image signals based on the flicker coefficient.

摘要翻译： 线路积分单元31对水平线的数字图像信号进行积分。 采样单元32将构成一帧的至少一个但不是全部的水平线中的每一个作为与第二行积分的第一行进行采样。 存储器33存储对应于多个帧的获取的第二行积分。 屏幕间DFT单元34对由采样单元32b最新采样的第二行积分和存储在存储器33中的第二行积分构成的第二行积分序列进行离散傅立叶变换。闪烁提取单元35提取闪烁分量 基于离散傅里叶变换的结果。 闪烁系数计算单元36基于该信息计算闪烁系数。 校正操作单元37基于闪烁系数执行减少图像信号上的闪烁分量的操作。

公开(公告)号：US20100021129A1

公开(公告)日：2010-01-28

申请号：US12524952

申请日：2008-12-02

申请人： Tadanori Tezuka ,  Ikuo Fuchigami

发明人： Tadanori Tezuka ,  Ikuo Fuchigami

IPC分类号： H04N5/91

CPC分类号： H04N5/85 ,  G11B27/034 ,  G11B2020/10537 ,  G11B2220/2562 ,  H04N5/765 ,  H04N5/775 ,  H04N5/781 ,  H04N9/7921 ,  H04N9/8205

摘要： A disc device 11 stores (i) videos and still images, and (ii) for each video/still image, image information including an imaging time at which the video/still image was taken. A time management unit 16 manages a reference time in accordance with playback performed by a display control unit 14. When a reception unit receives, from a user, a switch instruction that causes an image switch to be performed, a search unit 17 refers to the image information pieces to search for (i) from among the videos, a video that was entirely/partially taken during time range Ta, and (ii) from among the still images, a still image that was taken during time range Tb, both of the time ranges Ta and Tb including the reference time. The playback control unit 14 performs the image switch based on a result of the search performed by the search unit 17.

摘要翻译： 光盘装置11存储（i）视频和静止图像，以及（ii）对于每个视频/静止图像，包括拍摄视频/静止图像的成像时间的图像信息。 时间管理单元16根据显示控制单元14执行的回放来管理参考时间。当接收单元从用户接收到执行图像切换的切换指令时，搜索单元17参考 搜索（i）视频中的图像信息，在时间范围Ta内完全/部分拍摄的视频，和（ii）从静止图像中获取在时间范围Tb期间拍摄的静止图像，两者 包括参考时间的时间范围Ta和Tb。 播放控制单元14基于搜索单元17执行的搜索结果执行图像切换。

公开(公告)号：US20100013930A1

公开(公告)日：2010-01-21

申请号：US12438785

申请日：2007-10-04

申请人： Masatoshi Matsuo ,  Ikuo Fuchigami

发明人： Masatoshi Matsuo ,  Ikuo Fuchigami

IPC分类号： H04N7/18

CPC分类号： B60R1/00 ,  B60R2300/105 ,  B60R2300/302 ,  B60R2300/303 ,  B60R2300/605 ,  B60R2300/70 ,  B60R2300/8026 ,  B60R2300/804 ,  B60R2300/806 ,  G06T3/0075 ,  G06T11/00 ,  H04N7/181

摘要： A video display apparatus for displaying on a monitor videos captured using capturing sections is provided. The capturing sections (101a-101n) capture the periphery of a vehicle. Reference position setting sections (102a-102n) set a focus monitor position as a video construction reference position. A steering angle detecting section (108) detects a steering angle caused by a driver's operation. The video construction reference position is corrected based on the steering angle. After video construction reference position information adding sections (103a-103n) add video construction reference position information to the videos, a video constructing section (104) combines the videos of the capturing sections so that all of the video construction reference positions overlap each other. A display video generating section (106) clips a video based on a display area specified by a display area specifying section (105). A video display section (107) outputs the clipped video to the monitor for display.

摘要翻译： 提供一种用于在监视器上显示使用捕获部分捕获的视频的视频显示装置。 捕获部（101a-101n）捕获车辆的周边。 基准位置设定部（102a-102n）将焦点监视位置设定为视频构造基准位置。 转向角检测部（108）检测由驾驶员的操作引起的转向角。 基于转向角校正视频构造基准位置。 在视频构建参考位置信息添加部分（103a-103n）之后，将视频构造参考位置信息添加到视频中，视频构建部分（104）组合拍摄部分的视频，使得所有视频构造参考位置彼此重叠。 显示视频生成部（106）基于由显示区域指定部（105）指定的显示区域来剪辑视频。 视频显示部分（107）将剪辑的视频输出到监视器以进行显示。

公开(公告)号：US20080219347A1

公开(公告)日：2008-09-11

申请号：US12043313

申请日：2008-03-06

申请人： Tsuyoshi Nakamura ,  Ikuo Fuchigami ,  Masao Hamada ,  Tadanori Tezuka

发明人： Tsuyoshi Nakamura ,  Ikuo Fuchigami ,  Masao Hamada ,  Tadanori Tezuka

IPC分类号： H04N7/26

CPC分类号： H04N19/51 ,  G11B27/005

摘要： Provided are a moving picture coding method, a moving picture decoding method, a moving picture coding device, and a moving picture decoding device, by which skipped display can be performed even in a moving picture decoding device having insufficient processing capability and increase of an amount of coded data can be suppressed. The moving picture coding method of coding a moving picture stream includes: generating a plurality of combined pictures each of which is generated by combining a plurality of pictures into a combined picture having the plurality of pictures as regions, the plurality of pictures being a part of a plurality of pictures included in the moving picture stream; coding the plurality of the combined pictures; and setting a slice boundary between the regions in the combined picture. In the coding, for each of the plurality of combined pictures, when the combined picture has one or more to-be-displayed region except at least one not-to-be-displayed region, the to-be-displayed region is coded with reference to only a to-be-displayed region of another combined picture. Here, the one ore more to-be-displayed region is to be displayed for skipped reproduction.

摘要翻译： 提供了一种运动图像编码方法，运动图像解码方法，运动图像编码装置和运动图像解码装置，即使在具有不足的处理能力和数量增加的运动图像解码装置中也可以进行跳过显示 可以抑制编码数据。 对运动图像流进行编码的运动图像编码方法包括：生成多个组合图像，每个组合图像通过将多个图像组合为具有多个图像的组合图像作为区域而生成，多个图像是 包括在所述运动图像流中的多个图像; 编码所述多个组合图像; 并在组合图像中的区域之间设置切片边界。 在编码中，对于多个组合图像中的每一个，当组合图像具有除了至少一个不被显示区域之外的一个或多个待显示区域时，将被显示区域用 仅提及另一组合图片的待显示区域。 这里，要显示一个多个待显示区域，以便跳过再现。

公开(公告)号：US20130215319A1

公开(公告)日：2013-08-22

申请号：US13882837

申请日：2012-04-17

申请人： Hiroto Tomita ,  Ikuo Fuchigami

发明人： Hiroto Tomita ,  Ikuo Fuchigami

IPC分类号： G06K9/62 ,  H04N5/232

CPC分类号： G06K9/6201 ,  G02B7/28 ,  G02B7/36 ,  G03B13/36 ,  G03B15/00 ,  G06K9/00228 ,  G06K9/00369 ,  G06K9/3216 ,  G06T1/00 ,  H04N5/232 ,  H04N5/23212 ,  H04N5/23219

摘要： The stability is improved with which focus control is performed by an image capture device that brings a face region image into focus according to the contrast method. A human detection circuit 3 performs a human image search by using a threshold value Thh1. A face detection circuit 2 performs a face image search by using a threshold value Thf1. When an entire body image region corresponding to an entire body of a person is detected and a face image region corresponding to a face of the same person is detected in the captured image through the human image search and the face image search, the face detection circuit 2 performs redetermination with respect to the face image region by using a threshold value Thf2. The redetermination by using the threshold value Thf21 has higher accuracy compared to the face image search by using the threshold value Thf1.

摘要翻译： 通过使用根据对比度方法使面部区域图像成为焦点的图像捕获装置进行聚焦控制，稳定性得到改善。 人检测电路3通过使用阈值Thh1进行人物图像搜索。 面部检测电路2通过使用阈值Thf1进行面部图像检索。 当检测到与人的整个身体相对应的整个身体图像区域，并且通过人体图像搜索和面部图像检索在捕获的图像中检测到与同一人的面部对应的面部图像区域时，脸部检测电路 2通过使用阈值Thf2对面部图像区域执行重新确定。 通过使用阈值Thf21，通过使用阈值Thf1，与面部图像搜索相比，重新确定具有更高的精度。

公开(公告)号：US08451345B2

公开(公告)日：2013-05-28

申请号：US13126561

申请日：2009-11-18

申请人： Ikuo Fuchigami

发明人： Ikuo Fuchigami

IPC分类号： H04N9/73

CPC分类号： H04N5/243 ,  H04N5/2357

摘要： Line integration unit 31 performs integration on digital image signals of a horizontal line. Sampling unit 32 samples, as a second line integral, a first line integral relating to each of at least one but not all of horizontal lines constituting one frame. Memory 33 stores acquired second line integrals corresponding to a plurality of frames. Inter-screen DFT unit 34 performs discrete Fourier transform on a second line integral sequence composed of a second line integral most recently sampled by the sampling unit 32b and the second line integrals stored in the memory 33. Flicker extraction unit 35 extracts information of flicker component based on a result of the discrete Fourier transform. Flicker coefficient calculation unit 36 calculates a flicker coefficient based on the information. Correction operation unit 37 performs operation of reducing the flicker component on the image signals based on the flicker coefficient.

摘要翻译： 线路积分单元31对水平线的数字图像信号进行积分。 采样单元32将构成一帧的至少一个但不是全部的水平线中的每一个作为与第二行积分的第一行进行采样。 存储器33存储对应于多个帧的获取的第二行积分。 屏幕间DFT单元34对由采样单元32b最近采样的第二行积分和存储在存储器33中的第二行积分组成的第二行积分序列进行离散傅里叶变换。闪烁提取单元35提取闪烁分量 基于离散傅里叶变换的结果。 闪烁系数计算单元36基于该信息计算闪烁系数。 校正操作单元37基于闪烁系数执行减少图像信号上的闪烁分量的操作。

公开(公告)号：US20120212641A1

公开(公告)日：2012-08-23

申请号：US13462930

申请日：2012-05-03

申请人： Tadanori TEZUKA ,  Ikuo Fuchigami

发明人： Tadanori TEZUKA ,  Ikuo Fuchigami

IPC分类号： H04N5/228

CPC分类号： H04N5/232 ,  H04N1/2145 ,  H04N1/215 ,  H04N5/23232 ,  H04N5/23245 ,  H04N5/23293 ,  H04N5/2621 ,  H04N5/772 ,  H04N9/8042 ,  H04N9/8047 ,  H04N9/8205 ,  H04N9/8227 ,  H04N9/8233 ,  H04N2101/00

摘要： An imaging device for supporting the user to generate satisfactory synthesized images, not missing photo opportunities. The imaging device generates a plurality of still images continuously and generates a synthesized image from the still images. While the shooting button is pressed to instruct continuous shooting, the imaging device performs a process in which it writes still image data, which is obtained by the continuous shooting, into a recording medium sequentially. After the continuous shooting ends, the imaging device enters a wait state to wait for a new instruction, and then upon receiving an instruction to start generating a synthesized image, it performs a process of generating a synthesized image by using a plurality of pieces of still image data stored in the recording medium.

摘要翻译： 一种成像装置，用于支持用户产生令人满意的合成图像，而不缺少照片机会。 成像装置连续地生成多个静止图像，并从静止图像生成合成图像。 当按下拍摄按钮来指示连续拍摄时，成像装置执行其中将通过连续拍摄获得的静止图像数据顺序地写入记录介质的处理。 在连续拍摄结束后，成像装置进入等待状态等待新指令，然后在接收到开始生成合成图像的指令时，通过使用多个静止图像来执行生成合成图像的处理 存储在记录介质中的图像数据。

公开(公告)号：US08249359B2

公开(公告)日：2012-08-21

申请号：US12377191

申请日：2008-04-11

申请人： Hiroto Tomita ,  Ikuo Fuchigami

发明人： Hiroto Tomita ,  Ikuo Fuchigami

IPC分类号： G06K9/62

CPC分类号： G06K9/00228

摘要： A detector detects a specified image in an input image. The detector includes an area determination unit for determining, in the input image, a detection target area in which the specified image potentially exists, a setting unit for setting positions of a plurality of matching target ranges substantially in the detection target area, each of the matching target ranges being a predetermined size, so that the matching target ranges cover the detection target area, and each matching target range overlaps a neighboring matching target range by a predetermined overlap width, and a matching unit for detecting the specified image by matching a portion of the input image encompassed by each matching target range set by the setting unit and a template image for detecting the specified image.

摘要翻译： 检测器检测输入图像中的指定图像。 检测器包括区域确定单元，用于在输入图像中确定可能存在指定图像的检测目标区域，设置单元，用于设置基本上在检测目标区域中的多个匹配目标范围的位置， 匹配目标范围是预定大小，使得匹配目标范围覆盖检测目标区域，并且每个匹配目标范围与相邻匹配目标范围重叠预定重叠宽度，以及匹配单元，用于通过匹配目标范围的一部分 由由设置单元设置的每个匹配目标范围包含的输入图像和用于检测指定图像的模板图像。