会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 31. 发明授权
    • Method of detecting and correcting a direction of image data and
document image filing system employing the same
    • 检测和校正图像数据的方向的方法以及采用该方法的文档图像归档系统
    • US5471549A
    • 1995-11-28
    • US235195
    • 1994-04-29
    • Yasuo KurosuHidefumi MasuzakiHiromichi Fujisawa
    • Yasuo KurosuHidefumi MasuzakiHiromichi Fujisawa
    • G06K9/20G06K9/32
    • G06K9/3208G06K2209/01
    • A document is scanned by an optical scanner to be converted into digital image data to be transmitted to an image memory. Characters are extracted from the digital image data one by one and features are extracted from each of the extracted characters. The features as standard patterns stored in a character recognition dictionary are read out to be subjected to 90 degree rotation, 180 degree rotation and 270 degree rotation by a feature rotator, thereby to produce standard patterns of the four directions. Similarity measures are individually calculated between the standard patterns of the four directions and the features of each extracted character, with respect to the four directions. This processing is repeated for a plurality of characters recorded in the document to select a direction having the maximum add value of the similarity measures, thereby to detect an input direction of the document. When the detected direction is not a proper direction, the counterclockwise rotation is given to the digital image data to correct the digital image data to the proper direction.
    • 文件由光学扫描器扫描以被转换成数字图像数据以发送到图像存储器。 从数字图像数据逐个提取字符,并且从每个提取的字符中提取特征。 读出存储在字符识别字典中的作为标准图案的特征,通过特征旋转器进行90度旋转,180度旋转和270度旋转,从而产生四个方向的标准图案。 相对于四个方向,在四个方向的标准图案和每个提取的字符的特征之间分别计算相似性度量。 针对记录在文档中的多个字符重复该处理,以选择具有相似性度量的最大附加值的方向,从而检测文档的输入方向。 当检测到的方向不是正确的方向时,逆时针旋转被给予数字图像数据以将数字图像数据校正到正确的方向。
    • 32. 发明授权
    • Rotation processing method of image and system therefor
    • 图像和系统的旋转处理方法
    • US4975977A
    • 1990-12-04
    • US442196
    • 1989-11-28
    • Yasuo KurosuHidefumi Masuzaki
    • Yasuo KurosuHidefumi Masuzaki
    • G09G5/36G06T3/60
    • G06T3/602
    • In an image data rotating system for applying rotation processing to input image data of a large picture plane, the image data is divided into rectangular small blocks and rotation processing is applied to the image data of each block. The rotation processing to the image data of each block is performed by first reducing in the x-axis direction and enlarging in the y-axis direction. Then the first oblique axis transformation is applied. Any error data between the real oblique axis and the image element found in the final raster of the corresponding region is held in a memory. Subsequently, the second oblique axis transformation in the y-axis direction is applied, and error data is held in similar manner. Error data held as the initial value for the rotation processing to the image data of the adjacent region block is used, thereby allowing the rotation of the image data of a large picture plane to an arbitrary angle to be executed at high speed.
    • 在用于对输入大图像平面的图像数据进行旋转处理的图像数据旋转系统中,将图像数据分割为矩形小块,并对每个块的图像数据进行旋转处理。 通过首先在x轴方向上减少并且在y轴方向上放大来执行对每个块的图像数据的旋转处理。 然后应用第一个斜轴变换。 实际倾斜轴与在相应区域的最终光栅中发现的图像元素之间的任何误差数据被保存在存储器中。 随后,施加y轴方向的第二倾斜轴变换,并以类似的方式保持误差数据。 使用作为对相邻区域块的图像数据的旋转处理的初始值保持的错误数据,从而允许将大图像的图像数据的旋转高速地执行任意角度。