公开(公告)号：US20150022580A1

公开(公告)日：2015-01-22

申请号：US14332232

申请日：2014-07-15

Applicant: CANON KABUSHIKI KAISHA

Inventor： Fumitaka Goto ,  Hidetsugu Kagawa ,  Senichi Saito ,  Nobutaka Miyake ,  Shinjiro Hori ,  Akitoshi Yamada ,  Ryosuke Iguchi ,  Tomokazu Ishikawa ,  Kouta Murasawa ,  Hiromitsu Akiba ,  Junichi Nakagawa

IPC: B41J29/393

CPC classification number: H04N1/6047 ,  B41J2/195 ,  B41J2/2142 ,  B41J2/2146 ,  H04N1/6036

Abstract: Because the ejection state of a printing element in a printing apparatus may vary at all times, an inspection item for inspecting a printed image may be set in consideration of the ejection state of the printing element. An inspection item for inspecting a printed image may be set based on information including a temperature characteristic of a printing head and a state of ink such as an elapsed time from the last ejection. This allows image inspection that matches with variations in the ejection state of the printing element.

Abstract translation: 因为打印装置中的打印元件的喷射状态总是可以改变，所以可以考虑打印元件的喷射状态来设置用于检查打印图像的检查项目。 用于检查打印图像的检查项目可以基于包括打印头的温度特性和诸如从最后喷射的经过时间的墨水状态的信息来设置。 这允许与打印元件的喷射状态的变化匹配的图像检查。

公开(公告)号：US08929681B2

公开(公告)日：2015-01-06

申请号：US14061040

申请日：2013-10-23

Applicant: Canon Kabushiki Kaisha

Inventor： Kiyoshi Umeda ,  Ryosuke Iguchi

IPC: G06T5/40 ,  H04N1/46 ,  H04N1/60

CPC classification number: G06T5/40 ,  G06T5/009 ,  H04N1/46 ,  H04N1/60 ,  H04N1/6025

Abstract: In an image processing apparatus for carrying out a plurality of correction processes on an input image, a multi-dimensional histogram of the input image is calculated, and a feature amount of the input image for which a specific correction process has been carried out is analyzed based on the multi-dimensional histogram. Then, based on the result of the analysis, correction parameters to be used for another correction process are calculated.

Abstract translation: 在用于对输入图像执行多个校正处理的图像处理装置中，计算输入图像的多维直方图，并且分析已经执行了特定校正处理的输入图像的特征量 基于多维直方图。 然后，基于分析结果，计算要用于另一校正处理的校正参数。

公开(公告)号：US20140198354A1

公开(公告)日：2014-07-17

申请号：US14148903

申请日：2014-01-07

Applicant: Canon Kabushiki Kaisha

Inventor： Hiromitsu Akiba ,  Tetsuya Suwa ,  Nobutaka Miyake ,  Ryosuke Iguchi ,  Yusuke Hashii

IPC: H04N1/04

CPC classification number: H04N1/6027 ,  H04N1/4074

Abstract: A frequency distribution of luminances is acquired from an image data acquired by optically reading a document. A frequency distribution of luminances of a background of the document is generated using the maximum frequency of the acquired frequency distribution as the maximum frequency of the frequency distribution of the luminances of the background of the document. A difference between the generated frequency distribution of the luminances of the background of the document and the acquired frequency distribution of the luminances is acquired, and the frequency distribution of the luminances of the background of the document is corrected based on the acquired difference.

Abstract translation: 从通过光学读取文档获得的图像数据中获取亮度的频率分布。 使用获取的频率分布的最大频率作为文档的背景的亮度的频率分布的最大频率来生成文档的背景的亮度的频率分布。 获取文档背景的亮度的发生频率分布与获取的亮度的频率分布之间的差异，并且基于所获取的差异来校正文档的背景的亮度的频率分布。

公开(公告)号：US20140139855A1

公开(公告)日：2014-05-22

申请号：US14076710

申请日：2013-11-11

Applicant: CANON KABUSHIKI KAISHA

Inventor： Ryosuke Iguchi ,  Nobutaka Miyake ,  Akitoshi Yamada ,  Mitsuhiro Ono ,  Fumitaka Goto ,  Hidetsugu Kagawa ,  Tomokazu Ishikawa ,  Junichi Nakagawa ,  Senichi Saito

IPC: H04N1/46

CPC classification number: H04N1/46 ,  H04N1/4015 ,  H04N1/4051

Abstract: One dither mask having a highest spacial frequency is selected from a plurality of dither masks. Next, a granularity is obtained with reference to a table based on the selected dither mask and an ejection amount level per area. Moreover, a difference in granularity between adjacent areas is calculated with respect to all of the areas. A maximum value is obtained out of the obtained differences in granularity, and then, the maximum difference in granularity is compared with a determination threshold. When the maximum difference in granularity is the threshold or greater, it is determined whether or not a dither mask having a spacial frequency lower than that of the selected dither mask is stored in a memory. When there are dither masks having lower spacial frequencies, a dither mask having a spacial frequency lower by one level than that of the selected dither mask is selected.

Abstract translation: 从多个抖动掩模中选择具有最高空间频率的一个抖动掩模。 接下来，参考基于所选择的抖动掩模的表和每个区域的喷射量水平获得颗粒度。 此外，相对于所有区域计算相邻区域之间的粒度差异。 从获得的粒度差中获得最大值，然后将最大粒度差与判定阈值进行比较。 当粒度上的最大差值为阈值或更大时，确定具有低于所选择的抖动掩模的空间频率的抖动掩模是否存储在存储器中。 当存在具有较低空间频率的抖动掩模时，选择具有比所选择的抖动掩模低的一个等级的空间频率的抖动掩模。

公开(公告)号：US20140139854A1

公开(公告)日：2014-05-22

申请号：US14076647

申请日：2013-11-11

Applicant: CANON KABUSHIKI KAISHA

Inventor： Senichi Saito ,  Akitoshi Yamada ,  Tomokazu Ishikawa ,  Fumitaka Goto ,  Nobutaka Miyake ,  Mitsuhiro Ono ,  Ryosuke Iguchi ,  Hidetsugu Kagawa ,  Junichi Nakagawa

IPC: H04N1/46

CPC classification number: H04N1/46 ,  H04N1/4015 ,  H04N1/4051

Abstract: Nozzles in a print head are arrayed in a density of 600 dpi. Moreover, a dither matrix has a size of 16 pixels×16 pixels in 600 dpi. The dither matrix is repeatedly used. In the meantime, each of rectangles represents an HS processing unit. WHS=3 pixels. As a consequence, the relationship of a least common multiple below is established in a nozzle array direction: 3×WD=16×WHS. In this case, the cycle of interference unevenness can be prolonged to the least common multiple between WD and WHS, that is, 48 pixels (3WD). In this manner, the size of the dither matrix is not an integral multiple of the HS processing unit width, so that the cycle of interference unevenness can be prolonged more than the size of the dither matrix. Thus, the interference unevenness can be hardly recognized.

Abstract translation: 打印头中的喷嘴以600dpi的密度排列。 此外，抖动矩阵在600dpi中具有16像素×16像素的大小。 重复使用抖动矩阵。 同时，每个矩形表示HS处理单元。 WHS = 3像素。 因此，在喷嘴排列方向上建立以下最小公倍数的关系：3×WD = 16×WHS。 在这种情况下，干扰不均匀的周期可以延长到WD和WHS之间的最小公倍数，即48像素（3WD）。 以这种方式，抖动矩阵的大小不是HS处理单元宽度的整数倍，使得干涉不均匀的周期可以比抖动矩阵的大小更长。 因此，难以识别干涉不均匀性。

公开(公告)号：US20140139853A1

公开(公告)日：2014-05-22

申请号：US14076622

申请日：2013-11-11

Applicant: Canon Kabushiki Kaisha

Inventor： Fumitaka Goto ,  Akitoshi Yamada ,  Tomokazu Ishikawa ,  Senichi Saito ,  Nobutaka Miyake ,  Mitsuhiro Ono ,  Ryosuke Iguchi ,  Hidetsugu Kagawa ,  Junichi Nakagawa

IPC: G06K15/02

CPC classification number: G06K15/02 ,  H04N1/405 ,  H04N1/52

Abstract: When an input image is shifted by 640 pixels from a test pattern with reference to the position of a nozzle, the remainder is obtained by dividing 640 pixels by pixels of the dither matrix in an x direction. For example, when the size of the dither matrix in the x direction is 256 pixels, the dither matrix is shifted by 128 pixels in a direction reverse to the x direction. In this manner, the phase of the dither matrix at the time of the quantization during test pattern printing matches the phase of the dither matrix at the time of the quantization during input image printing. Consequently, unevenness of the dither matrix at a position N becomes the same in both of the test pattern and the input image. The HS correction to density unevenness caused by the unevenness of the dither matrix becomes suitable for the input image.

Abstract translation: 当输入图像相对于喷嘴的位置从测试图案偏移了640个像素时，通过在x方向上划分抖动矩阵的640个像素来获得剩余部分。 例如，当x方向上的抖动矩阵的大小为256像素时，抖动矩阵沿与x方向相反的方向偏移128像素。 以这种方式，在测试图案打印期间的量化时的抖动矩阵的相位匹配输入图像打印期间的量化时的抖动矩阵的相位。 因此，在位置N处的抖动矩阵的不均匀性在测试图案和输入图像两者中都是相同的。 由抖动矩阵的不均匀度引起的对浓度不均匀性的HS校正变得适合于输入图像。

公开(公告)号：US20140044354A1

公开(公告)日：2014-02-13

申请号：US14061040

申请日：2013-10-23

Applicant: CANON KABUSHIKI KAISHA

Inventor： Kiyoshi Umeda ,  Ryosuke Iguchi

IPC: G06T5/40

CPC classification number: G06T5/40 ,  G06T5/009 ,  H04N1/46 ,  H04N1/60 ,  H04N1/6025

Abstract: In an image processing apparatus for carrying out a plurality of correction processes on an input image, a multi-dimensional histogram of the input image is calculated, and a feature amount of the input image for which a specific correction process has been carried out is analyzed based on the multi-dimensional histogram. Then, based on the result of the analysis, correction parameters to be used for another correction process are calculated.

Abstract translation: 在用于对输入图像执行多个校正处理的图像处理装置中，计算输入图像的多维直方图，并且分析已经执行了特定校正处理的输入图像的特征量 基于多维直方图。 然后，基于分析结果，计算要用于另一校正处理的校正参数。