摘要：

针对传统的基于RGB颜色空间的高动态范围成像算法存在计算量大和颜色失真的问题,提出一种基于L*a*b*颜色空间的高动态范围图像合成算法。将一组不同曝光量的序列图像从RGB颜色空间转换到L*a*b*颜色空间,采用Debevec-Malik算法在L*通道上处理以获取相机响应曲线,并对亮度分量进行合成,得到高动态范围图像的亮度分量。对a*和b* 2个色度通道分别执行加权平均合成,得到高动态范围图像的2个色度分量。对合成的高动态范围图像的亮度通道采用色调映射后转换为RGB颜色空间进行显示。实验结果表明,与基于RGB颜色空间和基于YUV颜色空间的合成算法相比,该算法合成的高动态范围图像质量较高,计算速度较快。

关键词: 图像处理, L*a*b*颜色空间, 高动态范围, 相机响应函数, 色调映射

Abstract:

In order to solve the problems of large computation and color distortion of high dynamic range imaging algorithm based on RGB color space,a high dynamic range imaging synthesis algorithm based on L*a*b* color space is proposed.A series of images with different exposures are converted from RGB color space to L*a*b* color space.Debevec-Malik algorithm is used to process the images on L* channel.The camera response curve is obtained and the luminance component is synthesized to obtain the luminance image of the high dynamic range image.Two colorimetric images with high dynamic range are obtained by weighted average synthesis of a* and b* chromaticity channels respectively.The luminance channel of the synthesized high dynamic range image is tonal mapped and converted to RGB color space for display.Experimental results show that,compared with the synthesis algorithm based on RGB color space and YUV color space,the high dynamic range image synthesized by this algorithm has higher quality and faster computation speed.

Key words: image processing, L*a*b* color space, high dynamic range, camera response function, tone mapping

中图分类号: 

• TP391.41