公开(公告)号：US06466693B1

公开(公告)日：2002-10-15

申请号：US09321345

申请日：1999-05-27

申请人： Makoto Otsu ,  Yasushi Adachi ,  Toshihiro Kanata

发明人： Makoto Otsu ,  Yasushi Adachi ,  Toshihiro Kanata

IPC分类号： G06K900

CPC分类号： G06K9/38 ,  G06K2209/01 ,  H04N1/40062 ,  H04N1/6072

摘要： The present invention is an image processing apparatus comprising: dividing means for dividing an image signal obtained by scanning an original document into predetermined local blocks and output a predetermined including a target pixel; a binary coding processing section which converts the image signal in the predetermined block into binary code information; means for calculating each area (dimension) of the binary values; means for calculating number of times that the binary values are changed; a threshold processing section for performing threshold processing by properly and selectively using a threshold of the maximum density difference in the predetermined area, each area (dimension) value after binary coding and number of counts as a parameter; and a judgement processing section for judging the image signal in the predetermined area.

摘要翻译： 本发明是一种图像处理装置，包括：分割装置，用于将通过将原始文档扫描获得的图像信号划分成预定的局部块并输出包括目标像素的预定值; 二进制编码处理部分，将预定块中的图像信号转换为二进制码信息; 用于计算二进制值的每个区域（维度）的装置; 用于计算二进制值改变次数的装置; 阈值处理部分，用于通过适当地选择性地使用预定区域中的最大浓度差的阈值，二进制编码之后的每个区域（维数）和计数数作为参数来执行阈值处理; 以及判断处理部，用于判断所述预定区域中的图像信号。

公开(公告)号：US06473202B1

公开(公告)日：2002-10-29

申请号：US09314327

申请日：1999-05-19

申请人： Toshihiro Kanata ,  Yasushi Adachi ,  Makoto Otsu

发明人： Toshihiro Kanata ,  Yasushi Adachi ,  Makoto Otsu

IPC分类号： G06K1502

CPC分类号： H04N1/40062 ,  G06K2215/0094 ,  H04N1/56

摘要： An image processing apparatus includes: a parameter computing means which computes the maximum density value, the summation of density differences, the average density and the maximum density difference for each pixel in a mask containing a target pixel; and a color determination counting circuit which makes appropriate, selective threshold comparisons based on the computed result from the parameter computing circuit to determine an area in a mask as being chromatic, achromatic or ‘the other’ area.

摘要翻译： 一种图像处理装置包括：参数计算装置，其计算包含目标像素的掩模中的每个像素的最大浓度值，密度差的总和，平均密度和最大浓度差; 以及颜色确定计数电路，其基于来自参数计算电路的计算结果进行适当的选择性阈值比较，以将掩模中的区域确定为有色，无色或“另一”区域。

公开(公告)号：US06556313B1

公开(公告)日：2003-04-29

申请号：US09405966

申请日：1999-09-27

申请人： William Ho Chang ,  Makoto Otsu

发明人： William Ho Chang ,  Makoto Otsu

IPC分类号： H04N158

CPC分类号： H04N1/58

摘要： A method for color misregistration detection. Input image data is buffered and transferred to vector space. An examination window for a current pixel is established and foreground and background pixels within that window are selected. The current pixel is examined to determine if it is in an edge of a scanned object. If the current pixel is in an edge, it is analyzed for color misregistration. The analysis includes an application of Stokes' theorem. If the results of the application of Stokes' theorem indicate that the color misregistration is above a threshold, the pixel is identified as having color misregistration.

摘要翻译： 一种颜色对准检测方法。 输入图像数据被缓冲并传输到向量空间。 建立当前像素的检查窗口，并选择该窗口内的前景和背景像素。 检查当前像素以确定其是否在扫描对象的边缘。 如果当前像素处于边缘，则会分析颜色不对齐。 分析包括斯托克斯定理的应用。 如果应用斯托克斯定理的结果表明颜色偏移高于阈值，则像素被识别为具有颜色不对齐。

公开(公告)号：US06859228B1

公开(公告)日：2005-02-22

申请号：US09419602

申请日：1999-10-18

申请人： William Ho Chang ,  Makoto Otsu

发明人： William Ho Chang ,  Makoto Otsu

IPC分类号： H04N1/58 ,  H04N9/093 ,  G06K9/00

CPC分类号： H04N1/58 ,  H04N9/093

摘要： A method for color misregistration detection. Input image data is buffered and transferred to vector space. An examination window for a current pixel is established and foreground and background pixels within that window are selected. The current pixel is examined to determine if it is in an edge of a scanned object. If the current pixel is in an edge, it is deemed to have color misregistration. For pixels that have been deemed to have color misregistration, a correction value is determined and then adjusted prior to being applied to the pixel value.

摘要翻译： 一种颜色对准检测方法。 输入图像数据被缓冲并传送到向量空间。 建立当前像素的检查窗口，并选择该窗口内的前景和背景像素。 检查当前像素以确定其是否在扫描对象的边缘。 如果当前像素处于边缘，则被认为具有颜色不对齐。 对于被认为具有颜色不对齐的像素，确定校正值，然后在应用于像素值之前进行调整。