基于多梯度的区域自适应去马赛克算法

Adaptive region demosaicking using multi-color gradient

针对Bayer格式的彩色滤波阵列,鉴于传统去马赛克算法存在图像纹理和边缘模糊的现象,提出了一种基于多梯度的区域自适应去马赛克算法。该算法利用区域水平和垂直方向的梯度组算子来判断区域平滑程度和插值方向,从而有效地恢复绿色分量,同时根据色度差恒定原理来还原另外两个丢失的颜色信息。实验结果显示,相对于传统插值算法提高了图像的峰值信噪比,锐化了图像的纹理和边缘,改善了图像的视觉质量。

For the CFA in Bayer format, general interpolation method may blur the image edge and texture and introduce color artifacts near edges. This paper put forward an adaptive region demosaicking using multi-color gradient restoration algorithm for color images in Bayer format to reduce the color artifacts. It first recovered green components according to the operators and the texture regions. Then reconstructed the chrominance channels based on the interpolated green components and the constant-hue rule. This experiment results show that this algorithm is able to improve the CPSNR, sharpen textnre and edge of the image and enhance image quality.