公开(公告)号：US20120098985A1

公开(公告)日：2012-04-26

申请号：US12912536

申请日：2010-10-26

申请人： Mark Robertson ,  Ming-Chang Liu ,  Yoshihiro Murakami ,  Toru Kurata ,  Yutaka Yoneda

发明人： Mark Robertson ,  Ming-Chang Liu ,  Yoshihiro Murakami ,  Toru Kurata ,  Yutaka Yoneda

IPC分类号： H04N5/14 ,  H04N5/225

CPC分类号： H04N5/145 ,  G06T7/262 ,  G06T2207/20021 ,  G06T2207/20056

摘要： A method and system to improve the performance of phase correlation motion estimation for low-bit-precision implementation are described herein. Phase correlation uses the Fast Fourier Transform (FFT) with operations with infinite-precision constants. Since physical implementations use finite-precision arithmetic, there is some loss in precision relative to the ideal infinite-precision case. In low-complexity implementations, it is desirable to use as few bits as possible, and if the precision is too low, the performance of traditional phase correlation suffers. A pre-processing technique is applied to the data prior to taking the FFT, which minimizes the negative effects of finite precision in the FFT and allows high quality results from phase correlation. The pre-processing step is a content-dependent contrast adjustment that maps the range of the input images' pixel values to the range of input values for the FFT. There is no post-processing required after the FFT to compensate for the pre-processing step.

摘要翻译： 本文描述了一种用于提高低位精度实现的相位运动估计性能的方法和系统。 相位相关使用具有无限精度常数的运算的快速傅里叶变换（FFT）。 由于物理实现使用有限精度算术，相对于理想的无限精度情况，精度有一定的损失。 在低复杂度实现中，期望尽可能少地使用位，如果精度太低，则传统相位相关的性能受损。 在采用FFT之前，将预处理技术应用于数据，这最小化了FFT中有限精度的负面影响，并允许相位相关的高质量结果。 预处理步骤是内容相关的对比度调整，其将输入图像的像素值的范围映射到用于FFT的输入值的范围。 在FFT之后不需要后处理来补偿预处理步骤。

公开(公告)号：US20110229056A1

公开(公告)日：2011-09-22

申请号：US12728025

申请日：2010-03-19

申请人： Mark A. Robertson ,  Ming-Chang Liu

发明人： Mark A. Robertson ,  Ming-Chang Liu

IPC分类号： G06K9/32

CPC分类号： G06T7/248

摘要： An image alignment method includes computationally efficient methods of achieving high-accuracy local motion estimates by using phase correlation. The method also estimates motion reliability that allows a generic robust model fitting algorithm to produce more accurate results while operating much more efficiently. One of three methods are used to determine sub-pel motion estimation with improved accuracy. Each of the sub-pel motion estimation methods uses phase correlation, and are based on fitting computationally efficient 2-D quadratic surfaces to a phase correlation surface. A pre-filter is applied which shapes the phase correlation surface to enable appropriate fitting to the quadratic surface. Bias is also compensated for prior to applying a sub-pel motion estimation method. The method also estimates the reliability of the sub-pel motion estimates determined using phase correlation.

摘要翻译： 图像对准方法包括通过使用相位相关来实现高精度局部运动估计的计算有效的方法。 该方法还估计运动可靠性，允许通用鲁棒模型拟合算法在更有效地运行时产生更准确的结果。 使用三种方法之一来确定提高精度的子像素运动估计。 每个子像素运动估计方法使用相位相关，并且基于将计算有效的2-D二次曲面拟合到相位相关表面。 应用预滤波器，其形成相位相关表面以使得能够适当地拟合到二次曲面。 在应用子像素运动估计方法之前，偏置也被补偿。 该方法还估计使用相位相关确定的子像素运动估计的可靠性。

公开(公告)号：US20110176013A1

公开(公告)日：2011-07-21

申请号：US12689823

申请日：2010-01-19

申请人： Mark Robertson ,  Ming-Chang Liu

发明人： Mark Robertson ,  Ming-Chang Liu

IPC分类号： H04N5/228 ,  H04N5/14

CPC分类号： H04N5/145

摘要： A method to estimate segmented motion uses phase correlation to identify local motion candidates and a region-growing algorithm to group small picture units into few distinct regions, each of which has its own motion according to optimal matching and grouping criteria. Phase correlation and region growing are combined which allows sharing of information. Using phase correlation to identify a small number of motion candidates allows the space of possible motions to be narrowed. The region growing uses efficient management of lists of matching criteria to avoid repetitively evaluating matching criteria.

摘要翻译： 估计分段运动的方法使用相位相关来识别局部运动候选，以及区域增长算法将小图像单元分组成几个不同的区域，每个区域根据最优匹配和分组标准具有其自己的运动。 相位相关和区域增长相结合，允许信息共享。 使用相位相关来识别少量运动候选可以使可能的运动的空间变窄。 该地区的增长使用匹配标准列表的有效管理，以避免重复评估匹配标准。

公开(公告)号：US20100321513A1

公开(公告)日：2010-12-23

申请号：US12486487

申请日：2009-06-17

申请人： Ximin Zhang ,  Ming-Chang Liu

发明人： Ximin Zhang ,  Ming-Chang Liu

IPC分类号： H04N5/228 ,  G06K9/48 ,  G06K9/34

CPC分类号： G06K9/00

摘要： Content adaptive detection of images having stand-out objects involves block variance-based detection and determining if an object includes a stand-out object. The images with a stand-out object are further processed to isolate an object of interest. The images without a detected stand-out object are further processed with a transition map-based detection method which includes generating a transition map. If an object portrait is determined from the transition map, then the image is further processed to isolate the object of interest.

摘要翻译： 具有突出对象的图像的内容自适应检测涉及基于块方差的检测，并且确定对象是否包括独立对象。 具有突出对象的图像被进一步处理以隔离感兴趣的对象。 使用基于过渡映射的检测方法进一步处理没有检测到的突出对象的图像，该方法包括生成转换图。 如果从转换图确定对象画像，则进一步处理图像以隔离感兴趣的对象。

公开(公告)号：US20100265313A1

公开(公告)日：2010-10-21

申请号：US12426085

申请日：2009-04-17

申请人： Ming-Chang Liu ,  Mark Robertson

发明人： Ming-Chang Liu ,  Mark Robertson

IPC分类号： H04N7/00

CPC分类号： H04N5/232 ,  G06T3/4038 ,  H04N5/23238 ,  H04N5/23248

摘要： Apparatus and method for automatically generating panoramic still photographs from a sequence of images collected during panning. Programming within the camera allows creating the panoramic image output from multiple captured stills and/or video frames without laborious user “stitching”. A sequence of images are captured under control of the camera which span a desired subject area being panned (in any direction) by the user. As the images are being captured, the programming assures that the edges of adjacent images in the sequence sufficiently overlap one another as the desired subject area is being panned, as well as controlling other necessary camera adjustments (e.g., maintaining fixed focus). A set of sequential overlapping image frames is collected and combined to create at least one panoramic still photograph. The user can preferably change settings to control how the images are put together into the panoramic image output.

摘要翻译： 用于在平移期间收集的图像序列中自动生成全景静态照片的装置和方法。 相机内的编程允许创建从多个捕获的静止图像和/或视频帧输出的全景图像，而不需要费力的用户“拼接”。 在照相机的控制下拍摄一系列图像，该照片跨越由用户平移（在任何方向上）的期望的主题区域。 随着图像被捕获，编程确保了序列中相邻图像的边缘彼此充分重叠，因为所需的对象区域被平移，并且控制其它必要的相机调整（例如，保持固定焦点）。 收集并组合一组顺序重叠的图像帧以创建至少一个全景静止照片。 用户可以优选地改变设置以控制如何将图像放在一起进入全景图像输出。

公开(公告)号：US20100253817A1

公开(公告)日：2010-10-07

申请号：US12418207

申请日：2009-04-03

申请人： Gazi Ali ,  Ming-Chang Liu ,  Ximin Zhang

发明人： Gazi Ali ,  Ming-Chang Liu ,  Ximin Zhang

IPC分类号： H04N5/335 ,  G06K9/00

CPC分类号： H04N9/045 ,  G06T3/4015 ,  H04N2209/045

摘要： A method and apparatus for forming a demosaiced image from a color-filter-array (“CFA”) image is provided. The CFA image comprises a first set of pixels colored according to a first (e.g., a green) color channel, a second set of pixels colored according to a second (e.g., a red) color channel and a third set of pixels colored according to a third (e.g., blue) color channel. The method may include obtaining an orientation map, which includes, for each pixel of the color-filter-array image, an indicator of orientation of an edge bounding such pixel. The method may further include interpolating the first color channel at the second and third sets of pixels as a function of the orientation map so as to form a fourth set of pixels. The method may also include interpolating the second color channel at the first and third sets of pixels as a function of the orientation map and the fourth set of pixels; and interpolating the third color channel at the first and second sets of pixels as a function of the orientation map and the fourth set of pixels.

摘要翻译： 提供了一种用于从彩色滤光片阵列（“CFA”）图像形成去马赛克图像的方法和装置。 CFA图像包括根据第一（例如，绿色）颜色通道着色的第一组像素，根据第二（例如，红色）颜色通道着色的第二组像素，以及根据第 第三（例如蓝色）彩色通道。 该方法可以包括获得方位图，其包括对于彩色滤波器阵列图像的每个像素的边界边界的方向的指示符。 该方法可以进一步包括根据取向图来插值第二和第三组像素处的第一颜色通道，以便形成第四组像素。 该方法还可以包括根据取向图和第四组像素来内插第一和第三组像素处的第二颜色通道; 以及作为所述取向图和所述第四组像素的函数，在所述第一和第二像素组处插入所述第三颜色通道。

公开(公告)号：US07738727B2

公开(公告)日：2010-06-15

申请号：US11464483

申请日：2006-08-14

申请人： Rei-Hong Chang ,  Ming-Chang Liu ,  Fang-Mei Lu

发明人： Rei-Hong Chang ,  Ming-Chang Liu ,  Fang-Mei Lu

IPC分类号： G06K9/36

CPC分类号： G06T5/009 ,  G06T2207/20182

摘要： Brightness values of image frames are collected and counted to form a brightness distribution. The brightness distribution of the image frames is compared with predetermined brightness distributions. Brightness values of all pixels of successive image frames are adjusted according to a result of the comparison, so as to improve image contrast.

摘要翻译： 收集并计数图像帧的亮度值以形成亮度分布。 将图像帧的亮度分布与预定亮度分布进行比较。 根据比较结果调整连续图像帧的所有像素的亮度值，以提高图像对比度。

公开(公告)号：US20100111435A1

公开(公告)日：2010-05-06

申请号：US12266551

申请日：2008-11-06

申请人： Ximin Zhang ,  Ming-Chang Liu

发明人： Ximin Zhang ,  Ming-Chang Liu

IPC分类号： G06K9/40

CPC分类号： H04N19/117 ,  H04N19/14 ,  H04N19/176 ,  H04N19/61 ,  H04N19/86

摘要： A video system includes: analyzing video data, having a block; performing a transition change detection for determining a spatial intensity transition within the block; performing a block-wise similarity measurement on the block in the video data for identifying a blocking artifact; and filtering with a two dimensional cross filter every pixel in the block for removing the blocking artifact.

摘要翻译： 视频系统包括：分析具有块的视频数据; 执行用于确定所述块内的空间强度转换的转变变化检测; 对所述视频数据中的块进行块式相似度测量，以识别块伪影; 并使用块中的每个像素对二维交叉滤波进行滤波，以消除块伪影。

公开(公告)号：US20090262800A1

公开(公告)日：2009-10-22

申请号：US12148582

申请日：2008-04-18

申请人： Ximin Zhang ,  Ming-Chang Liu

发明人： Ximin Zhang ,  Ming-Chang Liu

IPC分类号： H04N7/30

CPC分类号： H04N19/122 ,  H04N19/14 ,  H04N19/176 ,  H04N19/196 ,  H04N19/61

摘要： Low complexity edge detection and DCT type selection method to improve the visual quality of H.264/AVC encoded video sequence is described. Encoding-generated information is reused to detect an edge macroblock. Variance and Mean Absolute Difference (MAD) of one macroblock shows a certain relationship that is able to be used to differentiate the edge macroblock and the non-edge macroblock. Also, the variance difference of neighbor macroblocks provides a hint for edge existence. Then, a block-based edge detection method uses this information. To determine the DCT type for each block, the detected edges are differentiated as visual obvious edge, texture-like edge, soft edge and strong edge. 8×8 DCT is used for texture-like edges and the 4×4 DCT is used for all the other edges. The result is an efficient and accurate edge detection and transform selection method.

摘要翻译： 描述了低复杂度边缘检测和DCT类型选择方法，以提高H.264 / AVC编码视频序列的视觉质量。 编码生成的信息被重新用于检测边缘宏块。 一个宏块的方差和平均绝对差（MAD）表示能够用于区分边缘宏块和非边缘宏块的一定关系。 此外，相邻宏块的方差差异提供了边缘存在的提示。 然后，基于块的边缘检测方法使用该信息。 为了确定每个块的DCT类型，检测到的边缘被区分为视觉明显的边缘，纹理样边缘，软边缘和强边。 8x8 DCT用于纹理样边缘，4x4 DCT用于所有其他边。 结果是一种高效准确的边缘检测和变换选择方法。

公开(公告)号：US20090074058A1

公开(公告)日：2009-03-19

申请号：US11855888

申请日：2007-09-14

申请人： Ximin Zhang ,  Ming-Chang Liu

发明人： Ximin Zhang ,  Ming-Chang Liu

IPC分类号： H04N7/12

CPC分类号： H04N19/19 ,  H04N19/103 ,  H04N19/12 ,  H04N19/154 ,  H04N19/172

摘要： In one embodiment, a coding mode selection method is provided to improve the visual quality of an encoded video sequence. The coding mode is selected based on a human visual tolerance level. Picture data may be received for a video coding process. The picture data is then analyzed to determine human visual tolerance adjustment information. For example, parameters of a cost equation may be adjusted based on the human visual tolerance level, which may be a tolerance that is based on a distortion bound that the human visual system can tolerate. The picture data may be analyzed in places that are considered visually sensitive areas, such as trailing suspicious areas, stripping suspicious areas, picture boundary areas, and/or blocking suspicious areas. Depending on what kind of visually sensitive area is found in the picture data, a parameter in a cost equation may be adjusted based on different visual tolerance thresholds. The coding mode is then determined based on the cost.

摘要翻译： 在一个实施例中，提供编码模式选择方法以提高编码视频序列的视觉质量。 基于人的视觉容忍度来选择编码模式。 可以接收用于视频编码处理的图像数据。 然后分析图像数据以确定人类视觉容差调整信息。 例如，成本方程的参数可以基于人的视觉容忍度来调整，其可以是基于人类视觉系统可以容忍的失真约束的公差。 可以在被认为是视觉敏感区域的地方分析图像数据，例如可疑区域，剥离可疑区域，图像边界区域和/或阻止可疑区域。 根据在图像数据中找到什么样的视觉敏感区域，可以基于不同的视觉容差阈值来调整成本方程中的参数。 然后根据成本确定编码模式。