公开(公告)号：US20060285595A1

公开(公告)日：2006-12-21

申请号：US11432468

申请日：2006-05-12

申请人： Tae-hyeun Ha ,  Jae-seok Kim ,  Ji-yong Park ,  Won-jae Lee

发明人： Tae-hyeun Ha ,  Jae-seok Kim ,  Ji-yong Park ,  Won-jae Lee

IPC分类号： H04N11/02 ,  H04N7/12 ,  H04N11/04 ,  H04B1/66

CPC分类号： H04N19/573 ,  H04N13/111 ,  H04N13/239 ,  H04N13/30 ,  H04N19/102 ,  H04N19/139 ,  H04N19/172 ,  H04N19/182 ,  H04N19/503 ,  H04N19/51 ,  H04N19/597

摘要： A method of generating an intermediate image between a first frame and a second frame, using the first frame and a third frame successively received from a first camera and the second frame and a fourth frame successively received from a second camera. The intermediate image generating method includes: calculating and summing pixel value differences between the first frame and the third frame in units of a block; determining whether or not a center pixel of a block has motion; calculating current disparity vectors corresponding to motion vectors between the first frame and the second frame, and calculating previous disparity vectors corresponding to motion vectors between the third frame and the fourth frame; and generating the intermediate image using the current disparity vectors if the center pixel of the block has motion, and calculating an intermediate value of the previous disparity vectors and generating the intermediate image using the intermediate value of the previous disparity vectors if the center pixel of the block has no motion.

摘要翻译： 使用从第一相机连续接收的第一帧和第三帧以及从第二相机连续接收的第二帧和第四帧产生第一帧和第二帧之间的中间图像的方法。 中间图像生成方法包括：以块为单位计算和求和第一帧与第三帧之间的像素值差; 确定块的中心像素是否具有运动; 计算与第一帧和第二帧之间的运动矢量相对应的当前视差矢量，以及计算与第三帧和第四帧之间的运动矢量相对应的先前视差矢量; 以及如果所述块的中心像素具有运动，则使用所述当前视差矢量生成所述中间图像;以及如果所述前一视差向量的中心像素，则计算所述先前视差矢量的中间值并使用所述先前视差矢量的中间值生成所述中间图像 块没有动作。

公开(公告)号：US20050030423A1

公开(公告)日：2005-02-10

申请号：US10879724

申请日：2004-06-30

申请人： Tae-hyeun Ha ,  Jae-seok Kim ,  Jun-ho Sung ,  Seong-joo Lee

发明人： Tae-hyeun Ha ,  Jae-seok Kim ,  Jun-ho Sung ,  Seong-joo Lee

IPC分类号： H04N5/44 ,  H04N7/01 ,  H04N11/20

CPC分类号： H04N7/012 ,  H04N7/014

摘要： There is disclosed an adaptive de-interlacing method and apparatus based on a phase corrected field of an input interlaced picture signal, and a recording medium storing computer programs for executing the method. The method includes: generating a phase corrected kth field by compensating a phase of the kth field of the input interlaced picture signal by 1 line frequency; calculating a first motion vector between the phase corrected kth field and the (k−1)th field of the interlaced picture signal; calculating a second motion vector between the (k−1)th field and the (k+1)th field of the interlaced picture signal; and determining whether there is motion of the input picture based on the first and second motion vector values, and generating a kth progressive frame based on the result and the kth field. Thus, a serration phenomenon is minimized after de-interlacing by more exactly detecting the motion information of a corresponding inserted field.

摘要翻译： 公开了一种基于输入隔行图像信号的相位校正场的自适应去隔行方法和装置，以及存储用于执行该方法的计算机程序的记录介质。 该方法包括：通过将输入的隔行图像信号的第k个场的相位补偿1行频率来产生相位校正的第k个场; 计算所述隔行图像信号的相位校正后的第k场和第（k-1）场之间的第一运动矢量; 计算隔行图像信号的第（k-1）场和第（k + 1）场之间的第二运动矢量; 以及基于所述第一和第二运动矢量值确定是否存在所述输入图像的运动，以及基于所述结果和所述第k个场生成第k个逐行帧。 因此，通过更准确地检测相应的插入场的运动信息，在去隔行后，锯齿现象被最小化。

公开(公告)号：US07324430B2

公开(公告)日：2008-01-29

申请号：US10695950

申请日：2003-10-30

申请人： Tae-hyeun Ha ,  Jae-seok Kim ,  Seong-joo Lee

发明人： Tae-hyeun Ha ,  Jae-seok Kim ,  Seong-joo Lee

IPC分类号： H04J11/00

CPC分类号： H04L27/2662 ,  H04Q11/02

摘要： An orthogonal frequency division multiplexing (OFDM)-based synchronization detection apparatus including a 2n level quantizing unit quantizing received data samples into levels of 2n, where n is an integer greater than or equal to zero (0), and a delaying unit delaying the data samples quantized through the 2n level quantizing unit by a predetermined number of clocks and outputting data indicative thereof. A shifting unit shifts the output data samples of the 2n level quantizing unit by an amount corresponding to an exponent of the output data of the delaying unit and a peak detecting unit detects a peak value from sums of outputs from the shifting unit.

摘要翻译： 一种基于正交频分复用（OFDM）的同步检测装置，包括将接收的数据样本量化为2×n的电平的2阶电平量化单元，其中n是更大的整数 以及延迟单元，通过预定数量的时钟延​​迟通过2＆lt; n＆gt;级量化单元量化的数据样本并输出表示其的数据。 移位单元将2＆lt; n＆gt;电平量化单元的输出数据样本移位一个与延迟单元的输出数据的指数相对应的量，并且峰值检测单元从输出的和来检测峰值 从换档单元。

公开(公告)号：US07286187B2

公开(公告)日：2007-10-23

申请号：US10879724

申请日：2004-06-30

申请人： Tae-hyeun Ha ,  Jae-seok Kim ,  Jun-ho Sung ,  Seong-joo Lee

发明人： Tae-hyeun Ha ,  Jae-seok Kim ,  Jun-ho Sung ,  Seong-joo Lee

IPC分类号： H04N11/00

CPC分类号： H04N7/012 ,  H04N7/014

摘要： There is disclosed an adaptive de-interlacing method and apparatus based on a phase corrected field of an input interlaced picture signal, and a recording medium storing computer programs for executing the method. The method includes: generating a phase corrected kth field by compensating a phase of the kth field of the input interlaced picture signal by 1 line frequency; calculating a first motion vector between the phase corrected kth field and the (k−1)th field of the interlaced picture signal; calculating a second motion vector between the (k−1)th field and the (k+1)th field of the interlaced picture signal; and determining whether there is motion of the input picture based on the first and second motion vector values, and generating a kth progressive frame based on the result and the kth field. Thus, a serration phenomenon is minimized after de-interlacing by more exactly detecting the motion information of a corresponding inserted field.

摘要翻译： 公开了一种基于输入隔行图像信号的相位校正场的自适应去隔行方法和装置，以及存储用于执行该方法的计算机程序的记录介质。 该方法包括：通过将输入的隔行扫描图像信号的第k个第场的相位补偿1行频率来产生相位校正的第k个场; 计算相位校正的第k个场和第（k-1）个第个隔行扫描图像信号的第一运动矢量; 计算所述隔行图像信号的第（k-1）^{场和第（k + 1）场之间的第二运动矢量; 以及基于所述第一和第二运动矢量值确定是否存在所述输入图像的运动，以及基于所述结果和所述第k个字段来生成第k个逐行帧。 因此，通过更准确地检测相应的插入场的运动信息，在去隔行后，锯齿现象被最小化。}

公开(公告)号：US07263131B2

公开(公告)日：2007-08-28

申请号：US10702034

申请日：2003-11-06

申请人： Tae-hyeun Ha ,  Jae-seok Kim ,  Seong-joo Lee

发明人： Tae-hyeun Ha ,  Jae-seok Kim ,  Seong-joo Lee

IPC分类号： H04K1/10

CPC分类号： H04L27/2656 ,  H04L27/2675

摘要： An orthogonal frequency division multiplexing (OFDM)-based synchronization detection apparatus includes an input register that stores data, a shifter that shifts the data from the input register based on the exponent of a quantized correlation coefficient, an adder that sums the shifted data, and a peak detector that determines the peak value from among the summed shifted data. The OFDM-based synchronization detection method involves quantizing correlation coefficients for synchronization detection into 2n-level quantized correlation coefficients (n is an integer not less than 0), shifting input data using the 2n-level quantized correlation coefficients, and detecting synchronization using a shifting result.

摘要翻译： 基于正交频分复用（OFDM）的同步检测装置包括存储数据的输入寄存器，基于量化相关系数的指数将来自输入寄存器的数据移位的移位器，对移位数据求和的加法器和 一个峰值检测器，用于从求和的移位数据中确定峰值。 基于OFDM的同步检测方法包括将用于同步检测的相关系数量化为2个低级量化相关系数（n是不小于0的整数），使用2≤n移位输入数据 级量化相关系数，并使用移位结果检测同步。

公开(公告)号：US20060062306A1

公开(公告)日：2006-03-23

申请号：US11171344

申请日：2005-07-01

申请人： Tae-hyeun Ha ,  Jae-seok Kim

发明人： Tae-hyeun Ha ,  Jae-seok Kim

IPC分类号： H04N11/02 ,  H04N7/12 ,  H04N11/04 ,  H04B1/66

CPC分类号： H04N19/577 ,  H04N19/51 ,  H04N19/517 ,  H04N19/53 ,  H04N19/533 ,  H04N19/56 ,  H04N19/57 ,  H04N19/583

摘要： Provided are a method of motion estimation based on hybrid block matching using overlapping blocks in a video encoder, and an apparatus for converting a frame rate using the method. The method includes dividing a current frame into blocks and performing a full search algorithm on blocks sampled from the divided blocks, allocating motion vectors (fMV) obtained through linear interpolation to non-sampled blocks in the current frame based on a motion vector (iMV) obtained through the full search, and performing a fast search algorithm using the motion vector obtained through linear interpolation as a search starting point.

摘要翻译： 提供了一种基于使用视频编码器中的重叠块的混合块匹配的运动估计方法以及使用该方法转换帧速率的装置。 该方法包括将当前帧划分成块，并对从分割块采样的块执行全搜索算法，基于运动矢量（iMV）向当前帧中的未采样块分配运动矢量（fMV） 通过全搜索获得，并且使用通过线性插值获得的运动矢量作为搜索起点执行快速搜索算法。

公开(公告)号：US08149305B2

公开(公告)日：2012-04-03

申请号：US11874441

申请日：2007-10-18

申请人： Seung-gi Shin ,  Jae-seok Kim ,  Dong-wook Kim ,  Hyung-han Lee ,  Won-jae Lee ,  Jee-hoon An

发明人： Seung-gi Shin ,  Jae-seok Kim ,  Dong-wook Kim ,  Hyung-han Lee ,  Won-jae Lee ,  Jee-hoon An

IPC分类号： H04N9/64

CPC分类号： H04N5/3675

摘要： A method and apparatus to detect a dead pixel in an image sensor having a pixel array to generate a plurality of image frames, and a method and apparatus to capture an image from the image sensor detect a dead pixel in the image sensor. A change of a scene of the image frames is sensed, and a detection operation to detect a dead pixel in the pixel array is performed if the change of the scene is sensed.

摘要翻译： 一种用于检测具有像素阵列以生成多个图像帧的图像传感器中的死像素的方法和装置，以及用于从图像传感器捕获图像的方法和装置，其检测图像传感器中的死像素。 感测到图像帧的场景的变化，并且如果感测到场景的改变，则执行用于检测像素阵列中的死像素的检测操作。

公开(公告)号：US20080239114A1

公开(公告)日：2008-10-02

申请号：US11874441

申请日：2007-10-18

申请人： Seung-gi Shin ,  Jae-seok Kim ,  Dong-wook Kim ,  Hyung-han Lee ,  Won-jae Lee ,  Jee-hoon An

发明人： Seung-gi Shin ,  Jae-seok Kim ,  Dong-wook Kim ,  Hyung-han Lee ,  Won-jae Lee ,  Jee-hoon An

IPC分类号： H04N9/64

CPC分类号： H04N5/3675

摘要： A method and apparatus to detect a dead pixel in an image sensor having a pixel array to generate a plurality of image frames, and a method and apparatus to capture an image from the image sensor detect a dead pixel in the image sensor. A change of a scene of the image frames is sensed, and a detection operation to detect a dead pixel in the pixel array is performed if the change of the scene is sensed.

摘要翻译： 一种用于检测具有像素阵列以生成多个图像帧的图像传感器中的死像素的方法和装置，以及用于从图像传感器捕获图像的方法和装置，其检测图像传感器中的死像素。 感测到图像帧的场景的改变，并且如果感测到场景的改变，则执行用于检测像素阵列中的死像素的检测操作。

公开(公告)号：US20050057645A1

公开(公告)日：2005-03-17

申请号：US10910809

申请日：2004-08-04

申请人： Jun-ho Sung ,  Sung-sik Kim ,  Jae-seok Kim

发明人： Jun-ho Sung ,  Sung-sik Kim ,  Jae-seok Kim

IPC分类号： H04N13/04 ,  G06F3/14 ,  H04N13/00 ,  H04N15/00

CPC分类号： H04N13/156 ,  H04N13/161 ,  H04N13/243 ,  H04N13/282 ,  H04N13/349 ,  H04N19/597

摘要： A 3-dimensional (3-D) image signal processing method and apparatus. The 3-D image signal processing apparatus converts a multi-visual-point 2-D image signal into a signal suited to a 3-dimensional image display apparatus. The 3-D image signal processing apparatus includes: a video decoder converting an analog image signal received from an image source arranged in an m*n space-division mode into a digital image signal; a 3-D image processing unit performing 3-D image processing of the digital image signal and generating an output signal suited to a 3-D image display apparatus; a memory storing the digital image signal; and a controller generating a memory-write-address and a memory-read-address of the memory and controlling write/read order of the digital image signal for scaling and vertical/horizontal multiplexing. According to the 3-D image signal processing apparatus of the present invention, a 3-D display interface apparatus capable of being applied to a plurality of multi-visual-point modes is realized.

摘要翻译： 3维（3-D）图像信号处理方法和装置。 3-D图像信号处理装置将多视点2-D图像信号转换为适合于三维图像显示装置的信号。 3-D图像信号处理装置包括：将从以m×n空间分割模式排列的图像源接收的模拟图像信号转换为数字图像信号的视频解码器; 3-D图像处理单元，对所述数字图像信号进行3-D图像处理，并生成适用于3-D图像显示装置的输出信号; 存储数字图像信号的存储器; 以及控制器，其生成存储器的存储器写入地址和存储器读取地址，并且控制用于缩放和垂直/水平复用的数字图像信号的写入/读取顺序。 根据本发明的3D图像信号处理装置，可以实现能够应用于多个多视点模式的3-D显示接口装置。

公开(公告)号：US07706621B2

公开(公告)日：2010-04-27

申请号：US10769777

申请日：2004-02-03

申请人： Jun-ho Sung ,  Jae-seok Kim ,  Seong-joo Lee

发明人： Jun-ho Sung ,  Jae-seok Kim ,  Seong-joo Lee

IPC分类号： G06K9/46

CPC分类号： G06K9/20 ,  G06K2209/40 ,  H04N19/593 ,  H04N19/597 ,  H04N19/61 ,  H04N19/96

摘要： A method and an apparatus to divide image blocks to improve the quality of intermediate images by splitting each macro image block (macro block) in left-eye and right-eye views into sub image blocks (sub blocks) according to quadtree disparity estimation, and then splitting each sub block into smaller sub blocks. The method includes setting a plurality of splitting threshold values for a macro block in an image frame and determining whether to split the macro block into sub blocks, and setting a plurality of splitting threshold values for each sub block and determining whether to split each sub block into smaller sub blocks.

摘要翻译： 通过根据四叉树视差估计将左眼和右眼视图中的每个宏图像块（宏块）分割为子图像块（子块），分割图像块以提高中间图像的质量的方法和装置，以及 然后将每个子块分割成更小的子块。 该方法包括为图像帧中的宏块设置多个分割阈值，并确定是否将宏块分割成子块，以及为每个子块设置多个分割阈值，并确定是否分割每个子块 变成更小的子块。