【24h】

Color Filtering Method for CFA Images Based on Gradient

机译:基于梯度的CFA图像色彩过滤方法

获取原文
获取原文并翻译 | 示例

摘要

Single-sensor digital cameras capture image by covering the sensor surface with a color filter array(CFA) such that each sensor pixel only samples one of three primary color values,three color is R(red),G(green) and B(blue). To render a full-color image, need an interpolation process commonly referred to as CFA demosaicking, is required to estimate the other two contributions for producing a full-color image. But, the noise in imaging sensors not only corrupts the color filter array, at the same time introduces artifacts during the color interpolation step and influence quality of images. In order to acquire high quality full-color images, adopt a sort of viable and effective interpolation algorithm based on gradient, at the time of removing the noise, reserve image border and detail information clearly.
机译:单传感器数码相机通过用滤色镜阵列(CFA)覆盖传感器表面来捕获图像,使得每个传感器像素仅采样三个基色值之一,三种颜色为R(红色),G(绿色)和B(蓝色) )。要渲染全色图像,需要进行插值处理(通常称为CFA去马赛克),以估计用于生成全色图像的其他两个贡献。但是,成像传感器中的噪声不仅破坏了滤色器阵列,同时在颜色插值步骤中引入了伪像,并影响了图像质量。为了获得高质量的全彩色图像,采用一种可行且有效的基于梯度的插值算法,在去除噪声时,要清晰地保留图像边界和细节信息。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号