公开(公告)号：US20130294708A1

公开(公告)日：2013-11-07

申请号：US13997284

申请日：2012-01-20

申请人： Tomohide Maeda ,  Kenji Tabei

发明人： Tomohide Maeda ,  Kenji Tabei

IPC分类号： G06T5/00

CPC分类号： G06T5/003 ,  G06T2207/20021 ,  H04N5/217 ,  H04N5/23264

摘要： An apparatus that can obtain a clear image by correcting blurs of respective object images even in a case where a plurality of objects with different blurs exists within an image. Local PSF estimating section estimates a PSF for each small region. PSF form identifying section identifies forms of the estimated PSFs of the small regions and groups local estimated PSFs such that small regions having the PSF forms with high similarity belong to the same group. Small region integrating section integrates adjacent small regions that are grouped in the same group. Due to this, object images blurring in a similar way can be accurately separated. As a result of this, a perspective separation of the objects and a separation of moving objects become possible from just one input image without using a plurality of cameras or additional devices (a distance sensor, etc.).

摘要翻译： 即使在图像中存在多个具有不同模糊的物体的情况下，也可以通过校正各个物体图像的模糊来获得清晰图像的装置。 局部PSF估计部分估计每个小区域的PSF。 PSF形式识别部分识别小区域的估计PSF的形式并且组合局部估计PSF，使得具有高相似性的形成PSF的小区域属于同一组。 小区域积分部分集成在相同组中的相邻小区域。 因此，可以精确地分离出以类似的方式模糊对象图像。 作为其结果，不使用多个照相机或附加装置（距离传感器等），只需要一个输入图像就可以使物体的透视分离和移动物体的分离。

公开(公告)号：US09609181B2

公开(公告)日：2017-03-28

申请号：US13992807

申请日：2011-12-20

申请人： Qinhe Wang ,  Kenji Tabei

发明人： Qinhe Wang ,  Kenji Tabei

IPC分类号： H04N5/217 ,  H04N5/232 ,  H04N5/262 ,  H04N1/387 ,  H04N1/407 ,  G06T3/40

CPC分类号： H04N5/2173 ,  G06T3/4053 ,  H04N1/387 ,  H04N1/407 ,  H04N5/23238 ,  H04N5/2628

摘要： An image signal processing apparatus and image signal processing method obtains a high super-resolution effect for distorted images captured by a specialized optical system, such as a 360-degree camera. A distortion-correction conversion-formula calculation section calculates a distortion correction conversion formula for correcting the nonlinear distortion. A distortion correction and interpolation processing section performs distortion correction and interpolation processing on each cropped region. An alignment conversion formula calculation section calculates an alignment conversion formula for alignment between each cropped image after distortion correction and interpolation processing is completed. A pixel coordinate conversion section rearranges pixel coordinates for each frame image before interpolation processing by using the distortion correction conversion formula and the alignment conversion formula corresponding to the cropped regions. An image synthesis section synthesizes the cropped images.

公开(公告)号：US20070222866A1

公开(公告)日：2007-09-27

申请号：US11568775

申请日：2005-10-11

申请人： Masayuki Serizawa ,  Kenji Tabei

发明人： Masayuki Serizawa ,  Kenji Tabei

IPC分类号： H04N5/232

CPC分类号： H04N5/232 ,  H04N5/205 ,  H04N5/208

摘要： It is an object of the present invention to provide a video signal processing device, which can produce a digital video signal, excellent in image quality, fully utilizing effective picture elements of the imaging element, without performing a special control of an imaging element driving unit. In accordance with a contour correction gain calculated based on a compression ratio detecting signal outputted from an image compressing unit 21, a contour correction unit 22 can carry out a contour correction on a compressed image. Accordingly, an NTSC/PAL encoder 24 can output a preferable video signal, which has been subjected to an appropriate contour correction. Moreover, a digital outputting unit 25 outputs an uncompressed video signal, and therefore can provide a preferable video signal without deteriorating the effective picture elements of an imaging element.

摘要翻译： 本发明的目的是提供一种视频信号处理装置，其可以产生图像质量优良的数字视频信号，充分利用成像元件的有效图像元素，而不进行成像元件驱动单元的特殊控制 。 轮廓校正单元22根据从图像压缩单元21输出的压缩比检测信号计算的轮廓校正增益，可以对压缩图像进行轮廓校正。 因此，NTSC / PAL编码器24可以输出经过适当轮廓校正的优选视频信号。 此外，数字输出单元25输出未压缩的视频信号，因此可以提供优选的视频信号，而不会劣化成像元件的有效图像元素。

公开(公告)号：US06593970B1

公开(公告)日：2003-07-15

申请号：US09187048

申请日：1998-11-06

申请人： Masayuki Serizawa ,  Kenji Tabei

发明人： Masayuki Serizawa ,  Kenji Tabei

IPC分类号： G03B700

CPC分类号： H04N5/235 ,  H04N9/045 ,  H04N2209/049

摘要： An imaging circuit generates a first video signal with a first exposure interval and a second video signal with a second exposure interval substantially at the same time, the second exposure interval being shorter than the first exposure interval, the first and second video signals respectively having first and second dynamic ranges which are different but continues. The first video signal is synchronized with the second video signal. An exposure ratio between the first and second exposure intervals is detected. A gain of the second video signal is adjusted according to the exposure ratio. A combined video signal is generated from the first and second video signals according to a mixing control signal indicative of a mixing ratio between first and second video signals and levels of the first and second video signals to have an expanded dynamic range such that the first dynamic range is connected to the second dynamic range with difference in gains of the first and second video signals adjusted for linearity. An edge enhancement signal may be gain-controlled or coring-controlled according to the mixing control signal or the exposure ratio. The similar apparatus and method for color signals are also disclosed. The output dynamic range for a display is limited by a non-linear process.

公开(公告)号：US20120213452A1

公开(公告)日：2012-08-23

申请号：US13372711

申请日：2012-02-14

申请人： Yoshiyuki Matsuyama ,  Kenji Tabei

发明人： Yoshiyuki Matsuyama ,  Kenji Tabei

IPC分类号： G06K9/32

CPC分类号： G06T3/4053

摘要： An image processing apparatus 100 selects one of multiple low-resolution images as a reference image candidate and determines transformation matrices for aligning the other aligning low-resolution images; conducts coordinate-transforming the other low-resolution images with the transformation matrices and plotting the reference image candidate and the coordinate-transformed low-resolution images on to a mapping image to generate a reconfigured image for reference image selection; gives a higher evaluation value to the reference image candidate as the number of pixels of the reference image candidate and the other low-resolution images plotted on the mapping image is larger. This image processing apparatus can select a reference image appropriate for generating a high-quality, high-resolution image from among multiple low-resolution images.

摘要翻译： 图像处理装置100选择多个低分辨率图像中的一个作为参考图像候选，并且确定用于对准其他对准低分辨率图像的变换矩阵; 使用变换矩阵进行坐标变换其他低分辨率图像，并将参考图像候选和坐标变换的低分辨率图像绘制到映射图像上，以生成用于参考图像选择的重新配置的图像; 当参考图像候选的像素数和映射图像上绘制的其他低分辨率图像的像素数量较大时，给予参考图像候选者更高的评估值。 该图像处理装置可以从多个低分辨率图像中选择适合于生成高质量，高分辨率图像的参考图像。

公开(公告)号：US07656431B2

公开(公告)日：2010-02-02

申请号：US10488271

申请日：2003-03-26

申请人： Masayuki Serizawa ,  Kenji Tabei

发明人： Masayuki Serizawa ,  Kenji Tabei

IPC分类号： H04N5/228 ,  H04N9/73 ,  H04N9/68

CPC分类号： H04N5/20 ,  H04N9/045

摘要： A video signal processing apparatus that reduces the degree of occurrences of false colors and of differing line concentration by performing knee processing properly, thereby providing desirable video signals. In the apparatus, at the occasion of generating luminance signals and color differences based on the result of knee processing, it is possible to properly generate a luminance signal which is derived from the addition result of pixels which are next to each other and generate a color difference signal which is derived from a difference of the pixels next to each other, achieved by performing the knee processing while keeping the signal level ratio at least between the pixels next to each other on video signals made up of color components arranged on a pixel by pixel basis. In this way, it is possible to keep a balance in hue of a color signal even after being subjected to knee processing.

摘要翻译： 一种视频信号处理装置，通过适当地进行膝盖处理来减少伪色发生的程度和不同的线浓度，从而提供期望的视频信号。 在该装置中，在基于膝部处理的结果生成亮度信号和色差的场合，可以适当地生成从彼此相邻的像素的相加结果导出并生成颜色的亮度信号 通过执行膝盖处理而实现的差分信号，同时保持信号电平比率至少在彼此相邻的像素之间，由由在像素上排列的颜色分量组成的视频信号上 像素基础。 以这种方式，即使在进行膝盖处理之后，也可以保持颜色信号的色调的平衡。

公开(公告)号：US20090046176A1

公开(公告)日：2009-02-19

申请号：US11577342

申请日：2005-10-14

申请人： Kenji Tabei

发明人： Kenji Tabei

IPC分类号： H04N5/262

CPC分类号： H04N5/232 ,  H04N5/23296

摘要： It is an object of the present invention to provide a video signal processing apparatus that can be reduced in production cost, and perform an electronic zooming function. In the video signal processing apparatus, a noise reduction of and a gain adjustment of the video signal from the imaging element 12 is performed by the analog preprocessing means 13. The video signal from the analog preprocessing means 13 is then converted into a digital signal by the A/D converter 14 in synchronization with the horizontal sync signal and its clock signal. The luminance and color difference signals are produced from the digital signal by the Y/C signal processing means 15, and delayed by the line delay means 16 on the basis of the period of the horizontal sync signal. The interpolating means 18 performs the vertical pixel interpolation process, on the basis of the vertical interpolation coefficient outputted by the zooming control means 17, by using the output signal of the Y/C signal processing means 15 and the output signal of the line delay means 16, and performs the horizontal pixel interpolation process, on the basis of the horizontal interpolation coefficient outputted by the zooming control means 17, by using pixels horizontally-adjacent to each other.

摘要翻译： 本发明的目的是提供一种视频信号处理设备，其可以降低生产成本并执行电子缩放功能。 在视频信号处理装置中，由模拟预处理装置13执行来自成像元件12的视频信号的噪声降低和增益调整。然后，来自模拟预处理装置13的视频信号被转换成数字信号 A / D转换器14与水平同步信号及其时钟信号同步。 亮度和色差信号由Y / C信号处理装置15从数字信号产生，并且由行延迟装置16根据水平同步信号的周期延迟。 内插装置18通过使用Y / C信号处理装置15的输出信号和行延迟装置的输出信号，根据变焦控制装置17输出的垂直内插系数，执行垂直像素内插处理 并且通过使用水平相邻的像素，基于由变焦控制装置17输出的水平内插系数，执行水平像素内插处理。

公开(公告)号：US06710818B1

公开(公告)日：2004-03-23

申请号：US09680216

申请日：2000-10-06

申请人： Misa Kasahara ,  Kenji Tabei ,  Makoto Sube

发明人： Misa Kasahara ,  Kenji Tabei ,  Makoto Sube

IPC分类号： H04N521

CPC分类号： H04N5/2357 ,  H04N5/235

摘要： A video signal including illumination flicker component is integrated at each of unit areas (horizontal lines) in a frame (field) of the video signal. The integrated level at each of the unit areas at the frame and the integrated level at the corresponding unit area of an adjacent frame are averaged. Dividing is effected between results of the averaging and integrating every unit area. It is judged whether flicker exists in the video signal by frequency-analyzing results of the dividing result at the unit areas. The unit area may be plural adjacent lines where flickering are negligible. The averaging circuit may be circulation type of or FIR filter. Threshold level for judging the flicker is changed according to a shutter speed control signal. Flicker compensation may be executed by controlling shutter speed or the AGC according to flicker judging result. A still condition at a block in a frame may be detected from the integration result at plural frames. When the block is judged to be still, the flicker is judged. An ac line frequency detection is also disclosed to detect the frequency of the ac line from a video signal generated under illumination including flicker. An imaging circuit may be provided to generate the video signal therein.

摘要翻译： 包括照明闪烁分量的视频信号被集成在视频信号的帧（场）中的每个单位区域（水平行）处。 在帧的每个单位区域和相邻帧的相应单位区域处的积分水平的积分水平被平均。 在每个单位面积的平均和积分结果之间进行分割。 通过频率分析在单位区域的分割结果的结果来判断视频信号中是否存在闪烁。 单位面积可以是多个相邻的线，其中闪烁是可忽略的。 平均电路可以是循环型或FIR滤波器。 用于判断闪烁的阈值水平根据快门速度控制信号而改变。 可以通过根据闪烁判断结果控制快门速度或AGC来执行闪烁补偿。 可以从多个帧的积分结果检测帧中的块处的静止条件。 当判断为静止时，判断闪烁。 还公开了一种交流线路频率检测，以从包括闪烁在内的照明下产生的视频信号检测交流线路的频率。 可以提供成像电路以在其中产生视​​频信号。

公开(公告)号：US5805216A

公开(公告)日：1998-09-08

申请号：US455153

申请日：1995-05-31

申请人： Kenji Tabei ,  Masato Nishizawa

发明人： Kenji Tabei ,  Masato Nishizawa

IPC分类号： G06T5/20 ,  H04N5/335 ,  H04N5/367 ,  H04N5/372 ,  H04N9/64

CPC分类号： H04N5/367 ,  G06T5/005 ,  G06T5/20

摘要： A defective pixel correction circuit corrects a defective pixel in a solid imaging device such as a CCD exactly and sufficiently. A boundary detection circuit calculates magnitudes of boundaries from signals of eight peripheral pixels taken in a pixel taking-in circuit and a boundary ordering circuit compares the calculated magnitudes of the boundaries with one another to order the magnitudes of the boundaries. An interpolation circuit produces an interpolation signal in accordance with an interpolation method determined by an interpolation method determining circuit on the basis of the order of the ordered boundaries to correct the defective pixel.

摘要翻译： 有缺陷的像素校正电路精确且充分地校正诸如CCD的固体成像装置中的缺陷像素。 边界检测电路根据在像素取入电路中拍摄的八个周边像素的信号来计算边界的大小，并且边界排序电路将计算出的边界的幅度相互比较以对边界的大小进行排序。 内插电路根据由内插方法确定电路确定的内插方法，根据有序边界的顺序来生成内插信号，以校正缺陷像素。

公开(公告)号：US20130258048A1

公开(公告)日：2013-10-03

申请号：US13992807

申请日：2011-12-20

申请人： Qinhe Wang ,  Kenji Tabei

发明人： Qinhe Wang ,  Kenji Tabei

IPC分类号： H04N5/217

CPC分类号： H04N5/2173 ,  G06T3/4053 ,  H04N1/387 ,  H04N1/407 ,  H04N5/23238 ,  H04N5/2628

摘要： An image signal processing apparatus and image signal processing method capable of obtaining a high super-resolution effect even for distorted images captured by a specialized optical system such as a 360-degree camera. Distortion-correction conversion-formula calculation section (123) calculates a distortion correction conversion formula for correcting the nonlinear distortion. Distortion correction and interpolation processing section (101) performs distortion correction and interpolation processing on each cropped region. Alignment conversion formula calculation section (104) calculates an alignment conversion formula for alignment between each cropped image after distortion correction and interpolation processing is completed. Pixel coordinate conversion section (111) rearranges pixel coordinates for each frame image before interpolation processing by using the distortion correction conversion formula and the alignment conversion formula corresponding to the cropped regions. Image synthesis section (112) synthesizes the cropped images, in which a plurality of frames have been rearranged, and which are obtained using the pixel coordinate conversion section (111).

摘要翻译： 一种图像信号处理装置和图像信号处理方法，即使对于诸如360度相机的专用光学系统捕获的失真图像，也能够获得高的超分辨率效果。 失真校正转换公式计算部（123）计算用于校正非线性失真的失真校正转换公式。 失真校正和内插处理部（101）对各裁剪区域进行失真校正和插值处理。 对准转换公式计算部（104）计算在畸变校正和插值处理完成后的每个裁剪图像之间的对准的对准转换公式。 像素坐标转换部（111）通过使用失真校正转换公式和对应于裁剪区域的对准转换公式，对插值处理前的每帧图像的像素坐标进行重新排列。 图像合成部分（112）合成已经重排了多个帧并且使用像素坐标转换部分（111）获得的裁剪图像。