一种亮度可控与细节保持的高动态范围图像色调映射方法

A Brightness-scaling and Detail-preserving Tone Mapping Method for High Dynamic Range Images

高动态范围(High dynamic range,HDR)图像通常需压缩其动态范围,以便于进行存储、传输、重现.本文提出一种具有亮度可控与细节保持特性的HDR图像的全局色调映射方法.该方法对HDR图像照度直方图进行裁剪与补偿,令色调映射后的低动态范围(Low dynamic range,LDR)图像仍能够保持原有的细节特性,同时利用概率模型估算出输出LDR图像的亮度与标准差,进而调整直方图亮度区域的分配,使得输出LDR图像的亮度接近用户设置的亮度,最后以分段直方图均衡的方法进行HDR色调映射处理.仿真结果表明,该方法能对HDR图像动态范围进行合理的压缩映射,输出的LDR图像的亮度可由用户控制或自适应选择,同时能保持图像的细节信息,令图像的主观视觉感受对比和谐.

For conventional storage, transfer, and display, high dynamic range （HDR） images always need to compress their dynamic ranges. This paper introduces a brightness-scaling and detail-preserving tone mapping method for HDR images. By clipping and compensating the luminance histogram of an HDR image, the method preserves the details after tone mapping. A probability model is also introduced to estimate the output brightness and its standard deviation. Then the histogram of the original image is divided into several regions to reduce the error between the expected brightness and the output brightness. Finally, the segmented histogram equalization is applied to each region for reproducing the low dynamic range image. Experiment results show that the new algorithm can successfully perform tone mapping for high dynamic range images. It can compress images according to a brightness given or set adaptively and preserve the details in images.