SINGLE ROW BASED DEFECTIVE PIXEL CORRECTION
    1.
    发明申请
    SINGLE ROW BASED DEFECTIVE PIXEL CORRECTION 有权
    基于单线的有缺陷的像素校正

    公开(公告)号:US20100103292A1

    公开(公告)日:2010-04-29

    申请号:US12260023

    申请日:2008-10-28

    IPC分类号: H04N9/64

    CPC分类号: H04N5/367

    摘要: An image sensor uses a single row of an array of pixels elements to determine whether a pixel is defective and to recover the defective pixel. The image sensor includes a “maximum of minimum” filter to remove a “black” pixel from a raw image. The image sensor also includes a “minimum of maximum” filter to remove a “white” pixel from the raw image.

    摘要翻译: 图像传感器使用单行像素元素阵列来确定像素是否有缺陷并恢复缺陷像素。 图像传感器包括从原始图像中去除“黑色”像素的“最小值”滤波器。 图像传感器还包括从原始图像中去除“白色”像素的“最小值”滤波器。

    Single row based defective pixel correction
    2.
    发明授权
    Single row based defective pixel correction 有权
    基于单行的缺陷像素校正

    公开(公告)号:US08164660B2

    公开(公告)日:2012-04-24

    申请号:US12260023

    申请日:2008-10-28

    IPC分类号: H04N9/64 H04N3/14

    CPC分类号: H04N5/367

    摘要: An image sensor uses a single row of an array of pixels elements to determine whether a pixel is defective and to recover the defective pixel. The image sensor includes a “maximum of minimum” filter to remove a “black” pixel from a raw image. The image sensor also includes a “minimum of maximum” filter to remove a “white” pixel from the raw image.

    摘要翻译: 图像传感器使用单行像素元素阵列来确定像素是否有缺陷并恢复缺陷像素。 图像传感器包括从原始图像中去除“黑色”像素的“最小值”滤波器。 图像传感器还包括从原始图像中去除“白色”像素的“最小值”滤波器。

    Method and apparatus for correcting for vignetting in an imaging system
    3.
    发明授权
    Method and apparatus for correcting for vignetting in an imaging system 有权
    用于在成像系统中校正渐晕的方法和装置

    公开(公告)号:US08823841B2

    公开(公告)日:2014-09-02

    申请号:US13528195

    申请日:2012-06-20

    CPC分类号: H04N5/3572 H04N9/045

    摘要: A method and apparatus for correcting for vignetting include associating each pixel in the two-dimensional array with a pair of polar coordinates referenced to a preselected origin pixel and partitioning the two-dimensional array of image pixels into a plurality of sectors. For each sector, the method includes computing an average R value, an average G value and an average B value; converting the average R value, the average G value and the average B value for each sector to logarithm space; comparing color gradients along a radial sector line to a gradient threshold; selecting gradients that do not exceed the threshold; using the selected gradients, estimating parameters of a model of a lens which produced the image; and, using the parameters, updating the model of the lens and correcting the image.

    摘要翻译: 用于校正渐晕的方法和装置包括将二维阵列中的每个像素与参考预选原始像素的一对极坐标相关联,并将图像像素的二维阵列分割成多个扇区。 对于每个扇区,该方法包括计算平均R值,平均G值和平均B值; 将每个扇区的平均R值,平均G值和平均B值转换为对数空间; 将沿着径向扇形线的颜色梯度与梯度阈值进行比较; 选择不超过阈值的梯度; 使用所选择的梯度,估计产生图像的透镜的模型的参数; 并使用参数,更新镜头的模型并校正图像。

    Method And Apparatus For Correcting For Vignetting In An Imaging System
    4.
    发明申请
    Method And Apparatus For Correcting For Vignetting In An Imaging System 有权
    在成像系统中校正晕影的方法和装置

    公开(公告)号:US20130342741A1

    公开(公告)日:2013-12-26

    申请号:US13528195

    申请日:2012-06-20

    IPC分类号: H04N9/083 G06T5/00

    CPC分类号: H04N5/3572 H04N9/045

    摘要: A method and apparatus for correcting for vignetting include associating each pixel in the two-dimensional array with a pair of polar coordinates referenced to a preselected origin pixel and partitioning the two-dimensional array of image pixels into a plurality of sectors. For each sector, the method includes computing an average R value, an average G value and an average B value; converting the average R value, the average G value and the average B value for each sector to logarithm space; comparing color gradients along a radial sector line to a gradient threshold; selecting gradients that do not exceed the threshold; using the selected gradients, estimating parameters of a model of a lens which produced the image; and, using the parameters, updating the model of the lens and correcting the image.

    摘要翻译: 用于校正渐晕的方法和装置包括将二维阵列中的每个像素与参考预选原始像素的一对极坐标相关联,并将图像像素的二维阵列划分成多个扇区。 对于每个扇区,该方法包括计算平均R值,平均G值和平均B值; 将每个扇区的平均R值,平均G值和平均B值转换为对数空间; 将沿着径向扇形线的颜色梯度与梯度阈值进行比较; 选择不超过阈值的梯度; 使用所选择的梯度,估计产生图像的透镜的模型的参数; 并使用参数,更新镜头的模型并校正图像。