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
    • 1. 发明授权
    • Print quality examining apparatus
    • 打印质量检查仪
    • US06373966B1
    • 2002-04-16
    • US09496846
    • 2000-02-02
    • Tomohiro FujiiNorifumi TasakaRieko Kitahara
    • Tomohiro FujiiNorifumi TasakaRieko Kitahara
    • G06K978
    • B41F33/00B41F33/0036G06K9/036G06T7/001G06T2207/30144
    • A printing quality examining apparatus in which image data printing paper is successively taken in by a camera of a detection unit. The image data are compared with a previously taken-in reference data to detect a printing defect so that a decision as to whether printing quality is good or bad is made. The image data is averaged in time to calculate an estimated data, and the estimated data is compared with the previously taken-in reference data. When the estimated data is normal, pixels are determined to be normal. When the estimated data is unusual, the image data from the detection unit is compared with the reference data at next step so that when the image data from the detection is normal, the pixels are determined to be normal and when the image data is unusual, the pixels are determined to be defective and the paper is discharged.
    • 一种打印质量检查装置,其中图像数据打印纸被检测单元的相机连续地取入。 将图像数据与先前获取的参考数据进行比较以检测打印缺陷,从而做出关于打印质量是好还是坏的决定。 图像数据在时间上被平均以计算估计数据,并且将估计数据与先前参考的参考数据进行比较。 当估计数据正常时,像素被确定为正常。 当估计数据异常时,将来自检测单元的图像数据与下一步骤的参考数据进行比较,使得当来自检测的图像数据正常时,像素被确定为正常,并且当图像数据异常时, 像素被确定为有缺陷并且纸张被排出。
    • 2. 发明授权
    • Method and apparatus for processing visual information
    • 用于处理视觉信息的方法和装置
    • US06466692B1
    • 2002-10-15
    • US09614545
    • 2000-07-12
    • Teruyoshi Washizawa
    • Teruyoshi Washizawa
    • G06K978
    • G06T5/20
    • A 2D image supplied from an image input unit including a wide view lens is sampled into a discrete form by an array sensor, and then mapped to a multi-resolution space by a 2D filter. The feature of the supplied image is detected, and then the mapped image is transformed to a local pattern about the detected feature, and then the coordinates of the position of the feature and the code word of the local pattern are formed into a set which is then encoded. The code is supplied to each cell of a stochastic automaton. The quantity of visual information is calculated in accordance with the quantity of mutual information between different cells of the stochastic automaton consisting of cells in blocks, the coordinates of the position of the feature and the distance from the feature to the optical axis so as to control the optical axis of the image input unit in such a manner that the quantity of visual information is maximized.
    • 从包括宽视角透镜的图像输入单元提供的2D图像由阵列传感器采样成离散形式,然后通过2D滤波器映射到多分辨率空间。 检测所提供的图像的特征,然后将映射的图像变换为关于检测到的特征的局部图案,然后将特征的位置坐标和局部图案的码字形成为 然后编码。 代码被提供给随机自动机的每个单元。 视觉信息量根据由块中的单元,特征位置的坐标以及从特征到光轴的距离的随机自动机的不同单元之间的相互信息量计算,以便控制 图像输入单元的光轴以使视觉信息量最大化的方式。
    • 3. 发明授权
    • Image recognition method and device and copier and scanner employing same
    • 图像识别方法和装置以及使用它的复印机和扫描仪
    • US06188787B1
    • 2001-02-13
    • US08826701
    • 1997-04-07
    • Kouichi OhmaeShinya SonodaIkuo KinoshitaHitoshi NakamuraJunji Hiraishi
    • Kouichi OhmaeShinya SonodaIkuo KinoshitaHitoshi NakamuraJunji Hiraishi
    • G06K978
    • H04N1/00846G03G21/046G07D7/00H04N1/00875
    • An image recognition method and device detects an object mark reliably even when the mark is defaced with a different color. RGB signals are transmitted to color extraction unit. The color extraction unit has a characteristic color extraction block to extract the characteristic color of the object mark. A binary image is generated from the extracted pixels which are the color of the mark and stored in storage device. A mark detection unit detects the location of the mark in the stored image data and an image extraction unit extracts the region where the mark is located and obtains a feature count. A matching unit compares the feature count with a reference pattern and finds the goodness of fit. A control unit subjects the goodness of fit value to threshold value processing to determine whether this is a nonreproducible document. If the mark has been defaced with black ink, the pixels constituting the mark are extracted by a likely color extraction block. A binary image is generated and stored in storage device. Thus, the defaced mark can undergo the same recognition processing as the normal mark, and both can be recognized.
    • 即使当标记被不同颜色污染时,图像识别方法和装置可靠地检测对象标记。 RGB信号被传送到颜色提取单元。 颜色提取单元具有特征颜色提取块,以提取对象标记的特征颜色。 从作为标记的颜色的提取像素生成二值图像并存储在存储装置中。 标记检测单元检测所存储的图像数据中的标记的位置,并且图像提取单元提取标记所在的区域并获得特征数。 匹配单元将特征计数与参考模式进行比较,并找到拟合优度。 控制单元对拟合值进行阈值处理,以确定这是否是不可再现的文档。 如果标记已被黑色油墨弄脏,则构成标记的像素由可能的颜色提取块提取。 二进制图像生成并存储在存储设备中。 因此,污损标记可以进行与正常标记相同的识别处理,并且都可以识别。