公开(公告)号：US5523792A

公开(公告)日：1996-06-04

申请号：US359153

申请日：1994-12-19

Applicant: John J. Ciardi ,  Keith R. Slavin

Inventor： John J. Ciardi ,  Keith R. Slavin

IPC: H04N5/073 ,  H04N7/01 ,  H04N11/04 ,  H04N9/455 ,  H03L7/00

CPC classification number: H04N7/0105 ,  H04N7/0102

Abstract: Composite video timing adjustments using digital resampling provides precise sub-pixel timing relative to a timing reference with a single video master clock. A digital resampler interpolates an input digital composite video signal as a function of a phase offset input which defines a fractional clock period of the video master clock derived from the timing reference. The sub-pixel offset digital composite video signal also may be adjusted by integer multiples of the video master clock period using a FIFO memory. The final timing adjusted digital composite video signal then is input to an analog reconstruction circuit to provide an output analog composite video signal that is precisely timed to the timing reference.

Abstract translation: 使用数字重采样的复合视频定时调整相对于具有单个视频主时钟的定时参考提供精确的子像素定时。 数字重采样器根据相位偏移输入内插输入数字复合视频信号，该相位偏移输入定义从时序参考导出的视频主时钟的分数时钟周期。 子像素偏移数字复合视频信号也可以使用FIFO存储器来调整视频主时钟周期的整数倍。 最后定时调整的数字复合视频信号然后被输入到模拟重建电路，以提供精确定时到定时参考的输出模拟复合视频信号。

公开(公告)号：US08788791B2

公开(公告)日：2014-07-22

申请号：US13345530

申请日：2012-01-06

Applicant: Keith R. Slavin

Inventor： Keith R. Slavin

IPC: G06F12/00 ,  G06F12/10 ,  G06F3/06

CPC classification number: G11C15/00 ,  G06F3/0628 ,  G06F12/1018 ,  G06F17/30982 ,  H04L45/00

Abstract: A comparand word is input to a plurality of hash circuits with each hash circuit responding to a different portion of the comparand word. The hash circuit outputs a hash signal which enables or pre-charges portions of a content addressable memory (CAM). The comparand word is also input to the CAM. The CAM compares the comparand word in the pre-charged portions of the CAM and outputs information responsive to the comparison. When Internet addresses are processed, the output information may be port information or an index for locating.

Abstract translation: 比较字被输入到多个散列电路，每个散列电路响应比较字的不同部分。 散列电路输出能够或预充电内容可寻址存储器（CAM）部分的散列信号。 比较字也输入到CAM。 CAM比较CAM的预充电部分中的比较字，并响应于比较输出信息。 当处理Internet地址时，输出信息可以是端口信息或用于定位的索引。

公开(公告)号：US08698908B2

公开(公告)日：2014-04-15

申请号：US12069669

申请日：2008-02-11

Applicant: Keith R. Slavin

Inventor： Keith R. Slavin

IPC: H04N5/228

CPC classification number: H04N5/357 ,  H04N5/23248 ,  H04N5/23267 ,  H04N5/3532

Abstract: A rolling shutter digital camera. Each photographic image of a given exposure duration is captured as a multi-frame burst of frames each having a shorter exposure duration to minimize motion blur and to reduce sensor noise by averaging. Each frame is quantized into swaths, captured sequentially by the rolling shutter. Swaths of the first frames are analyzed to select a set of best motion detection reference regions. Swaths of subsequent frames are analyzed versus only those regions, to reduce required computation, and are re-registered accordingly. Corresponding swaths of each frame are accumulated. The accumulator is normalized to the desired bit depth and written as the final image. Averaging of the multiple frames is improved by re-registering swaths rather than entire frames, because of the time delta caused by the rolling shutter. Computation is further reduced by selecting candidate points only along a key line of each swath of the first frame, and still further by pre-limiting the search by using a thumbnail version of the first frame to cull most candidate points. Thresholds are used to ensure adequate motion detection regions are used for each swath.

Abstract translation: 滚动快门数码相机。 给定曝光持续时间的每个摄影图像被捕获为具有较短曝光持续时间的帧的多帧突发，以最小化运动模糊并且通过平均来降低传感器噪声。 每个帧被量化成条，由滚动快门顺序捕获。 分析第一帧的幅面以选择一组最佳运动检测参考区域。 分析后续帧的幅面，而不仅仅是那些区域，以减少所需的计算，并相应地重新注册。 每个帧的相应条带被累积。 将累加器归一化为所需的位深度并写入最终图像。 由于由滚动快门引起的时间增量，可以通过重新排列条带而不是整个帧来改善多个帧的平均。 通过仅沿着第一帧的每个条带的键线选择候选点进一步减少计算，并且还通过使用第一帧的缩略图版本预先限制搜索以剔除大多数候选点。 阈值用于确保为每个条带使用足够的运动检测区域。

公开(公告)号：US08112578B2

公开(公告)日：2012-02-07

申请号：US10002461

申请日：2001-11-01

Applicant: Keith R. Slavin

Inventor： Keith R. Slavin

IPC: G06F13/00

CPC classification number: G11C15/00 ,  G06F3/0628 ,  G06F12/1018 ,  G06F17/30982 ,  H04L45/00

Abstract: A comparand word is input to a plurality of hash circuits, with each hash circuit responding to a different portion of the comparand word. The hash circuits output a hash signal which enables or precharges portions of a content addressable memory CAM. The comparand word is also input to the CAM. The CAM compares the comparand word in the precharged portions of the CAM and outputs information responsive to the comparison. When Internet addresses are processed, the output information is either port information or an index for locating port information.

Abstract translation: 比较字被输入到多个散列电路，每个散列电路响应比较字的不同部分。 散列电路输出可使内容可寻址存储器CAM的部分能够或预充电的散列信号。 比较字也输入到CAM。 CAM比较CAM的预充电部分中的比较字，并响应于比较输出信息。 当处理Internet地址时，输出信息是端口信息或用于定位端口信息的索引。

公开(公告)号：US06970170B2

公开(公告)日：2005-11-29

申请号：US10869554

申请日：2004-06-15

Applicant: Keith R. Slavin

Inventor： Keith R. Slavin

IPC: G06T3/40 ,  G09G5/00 ,  G09G5/36 ,  G06T11/00

CPC classification number: G06T3/4007 ,  G06T1/60 ,  G06T3/4023 ,  G09G5/006 ,  G09G5/363 ,  G09G2340/0414 ,  G09G2340/0421

Abstract: A resampling circuit and method where input sample values for samples arranged along a row of a source image are received by a row resampling circuit. The row resampling circuit calculates row output values which are provided to a column resampling circuit that calculates output sample values therefrom. The column resampling circuit includes a shift register that receives the row output values and shifts the row output values through the shift register as the row output samples are calculated. The shift register has a plurality of evenly spaced sample output terminals from which the row output values are sampled by an interpolation circuit for calculation of the output sample values.

Abstract translation: 一种重采样电路和方法，其中沿着源图像行排列的采样的输入采样值由行重采样电路接收。 行重采样电路计算提供给从其计算输出采样值的列重采样电路的行输出值。 列重采样电路包括移位寄存器，其接收行输出值，并且随着计算行输出采样而将行输出值移动通过移位寄存器。 移位寄存器具有多个均匀间隔的采样输出端子，通过用于计算输出采样值的内插电路对行输出值进行采样。

公开(公告)号：US06950048B1

公开(公告)日：2005-09-27

申请号：US10817477

申请日：2004-04-02

Applicant: Keith R. Slavin

Inventor： Keith R. Slavin

IPC: H03M1/20 ,  H03M1/06 ,  H03M1/08 ,  H03M1/10 ,  H03M1/12

CPC classification number: H03M1/0639 ,  H03M1/12

Abstract: A dither system for a quantizing device, such as a multi-stage pipelined analog-to-digital converter (ADC), derives a dither signal from a clock signal having a sample frequency, the dither signal having a frequency that is one-third of the sample frequency. The dither signal is easily converted to analog and added at the input of the quantizing device to an analog signal to be digitized. A cancellation signal circuit generates a cosine-wave signal from a digital version of the dither signal and programmable coefficients that are a function of amplitude and phase. The cosine-wave signal is combined with the digital output signal from the quantizing device to produce a corrected digital output signal having reduced quantization distortion.

Abstract translation: 用于量化装置的抖动系统，例如多级流水线模数转换器（ADC），从具有采样频率的时钟信号中导出抖动信号，抖动信号具有频率为三分之一的频率 采样频率。 抖动信号容易转换为模拟信号，并在量化装置的输入端加到待数字化的模拟信号上。 消除信号电路从抖动信号的数字版本和作为幅度和相位的函数的可编程系数产生余弦波信号。 余弦波信号与来自量化装置的数字输出信号组合，以产生具有降低的量化失真的校正数字输出信号。

公开(公告)号：US06911925B1

公开(公告)日：2005-06-28

申请号：US10817203

申请日：2004-04-02

Applicant: Keith R. Slavin

Inventor： Keith R. Slavin

IPC: H03F1/32 ,  H03M1/06 ,  H03M1/08 ,  H03M1/12 ,  H03M1/16 ,  H03M1/44 ,  H03M1/10

CPC classification number: H03M1/0626 ,  H03M1/0614 ,  H03M1/12 ,  H03M1/164 ,  H03M1/44

Abstract: A linearity compensation circuit provides cancellation of harmonic distortion introduced by a non-linear system into an input signal by providing a separate harmonic correction component for each harmonic of a fundamental frequency of the input signal. The harmonic correction components are summed with a delayed version of the input signal to produce a corrected input signal for input to the non-linear system. The separate harmonic correction components are generated by respective harmonic correction units, each having a programmable input filter, a Hilbert Transformer filter and a plurality of phase shifters. The output of each programmable input filter is input to the Hilbert Transformer filter and a compensating delay. The output from the Hilbert Transformer filter and the delayed version of the filtered input signal from the compensating delay are input to parallel phase shifters, and the respective outputs from the phase shifters are multiplied together to produce the separate harmonic correction component for each harmonic correction unit.

Abstract translation: 线性补偿电路通过为输入信号的基频的每个谐波提供单独的谐波校正分量来将由非线性系统引入的谐波失真消除为输入信号。 谐波校正分量与输入信号的延迟版本相加以产生用于输入到非线性系统的校正输入信号。 单独的谐波校正分量由相应的谐波校正单元产生，每个谐波校正单元具有可编程输入滤波器，希尔伯特变换器滤波器和多个移相器。 每个可编程输入滤波器的输出输入到希尔伯特变压器滤波器和补偿延迟。 来自希尔伯特变压器滤波器的输出和来自补偿延迟的经滤波的输入信号的延迟版本被输入到并联移相器，并且来自移相器的相应输出相乘以产生用于每个谐波校正单元的单独的谐波校正分量 。

公开(公告)号：US06088388A

公开(公告)日：2000-07-11

申请号：US931409

申请日：1997-09-16

Applicant: Keith R. Slavin

Inventor： Keith R. Slavin

IPC: H03H17/06 ,  H03H7/30

CPC classification number: H03H17/06

Abstract: A digital finite impulse response (FIR) filter for compensation of Nyquist slope in a pseudo-synchronous demodulator has a plurality of individual FIR filters coupled in series, having a bandpass filtered intermediate frequency video signal as an input and providing a broadband video signal as an output. The individual FIR filter amplitude responses represent each product term of the following polynomial:A(.omega.)=2(1+x)(1+x.sup.2) . . . (1+x.sup.2 (F-1))where F is the number of individual FIR filters in the series and x=(.omega..sub.m -.omega.)/.omega..sub.t, .omega..sub.m being the modulation frequency and .omega..sub.t being a deviation frequency. The register delay value for each of the individual FIR filters is obtained heuristically to achieve optimum quality as a compromise between performance and hardware complexity, and the number of individual FIR filters is determined by convolving each individual FIR filter with prior individual FIR filters in the series until the overall performance meets frequency response requirements.

Abstract translation: 用于补偿伪同步解调器中奈奎斯特斜率的数字有限脉冲响应（FIR）滤波器具有串联耦合的多个单独的FIR滤波器，其具有带通滤波的中频视频信号作为输入，并提供宽带视频信号作为 输出。 单个FIR滤波器幅度响应表示以下多项式的每个乘积项：A（ω）= 2（1 + x）（1 + x2）。 。 。 （1 + x2 + E，cir + EE（F-1））其中F是串联中的各个FIR滤波器的数量，x =（ω-ω-ω）/ω，t，ωm是调制频率，ω 作为偏差频率。 各个FIR滤波器的寄存器延迟值被初步获得，以获得最佳质量作为性能和硬件复杂度之间的折中，并且通过将每个单独的FIR滤波器与系列中的先前的单个FIR滤波器进行卷积来确定各个FIR滤波器的数量 直到整体性能符合频率响应要求。

公开(公告)号：US5612746A

公开(公告)日：1997-03-18

申请号：US419561

申请日：1995-04-10

Applicant: Keith R. Slavin

Inventor： Keith R. Slavin

IPC: H04N7/26 ,  H04N7/50 ,  H04N7/24

CPC classification number: H04N19/126 ,  H04N19/51 ,  H04N19/61 ,  H04N19/124

Abstract: Block matching for picture motion estimation uses offset quantized pixel values. The pixels of a reference block from a first frame of a video picture sequence are offset quantized to produce a pair of offset quantized reference pixel signals. The pixels of a search area from a next frame of the video picture sequence, corresponding to a current position of the reference block relative to the search area, the search area being larger than the reference block, are quantized to produce a quantized search pixel signal. The quantized search pixel signal is compared with each of the offset quantized reference pixel signals on a pixel by pixel basis to determine whether absolute equality exists between the respective pairs of pixel signals. The comparison results are combined so that for each quantized pixel of the search block that has absolute equality with either offset quantized reference pixel signal, a tally count is incremented. The process is repeated for each possible position of the reference block within the search area, and the portion of the search area having the greatest tally count is deemed to be the new position of the reference block in the second frame. From the new position appropriate motion estimation vectors may be determined.

Abstract translation: 图像运动估计的块匹配使用偏移量化像素值。 来自视频图像序列的第一帧的参考块的像素被偏移量化，以产生一对偏移量化参考像素信号。 对应于参考块相对于搜索区域的当前位置（搜索区域大于参考块）的视频图像序列的下一帧的搜索区域的像素被量化以产生量化的搜索像素信号 。 将量化的搜索像素信号逐个像素地与每个偏移量化参考像素信号进行比较，以确定在各个像素信号对之间是否存在绝对相等性。 比较结果被组合，使得对于与偏移量化参考像素信号绝对相等的搜索块的每个量化像素，递增计数。 对搜索区域内的参考块的每个可能的位置重复该过程，并且具有最大计数的搜索区域的部分被认为是第二帧中的参考块的新位置。 可以从新位置确定适当的运动估计向量。

公开(公告)号：US4758877A

公开(公告)日：1988-07-19

申请号：US30772

申请日：1987-03-24

Applicant: Keith R. Slavin

Inventor： Keith R. Slavin

IPC: H04N17/02 ,  H04N17/04 ,  H04N17/00

CPC classification number: H04N17/045

Abstract: A method for automatically finding color bars within a television video signal checks for a black burst at the end of a horizontal video line to determine the appropriate color bar standard for testing. Then a reference color burst, such as the yellow color burst, is searched for in the appropriate location within the horizontal line. Related color burst locations are demodulated to produce related color angles which are checked for relationship with the reference color burst. Also the luminance amplitudes for each color burst are compared to each other to determine whether a correct amplitude relationship is present. Each horizontal line of a picture frame is tested until either the color bars are found or all the horizontal lines have been tested. The found color bars are displayed in an appropriate format together with the line number within the picture frame, or a message indicating no color bars is displayed.