基于多曝光的高动态图像合成的噪声处理

Noise Processing of High Dynamic Image Synthesis Based on Multiple Exposure

针对在高动态范围图像合成的过程中有噪声影响图像的质量这一问题,采取一种基于多曝光图像的高动态范围图像合成降噪算法。通过对各曝光图像的灰度数据进行提取、整理、分析,能合成代表原始场景光线分布的亮度图像。通过分析噪声对高动态范围图像合成质量的影响,提出在图像合成前将图像中含有的噪声进行处理。根据光子散粒噪声变化的特点,将图像混有的噪声问题转化为求解一个多曝光图像序列组的平均值问题,合成的图像视觉效果与真实图像极为接近。

Aimingat the problem that the quality of image is affected by noise in the process of high dynamic range image synthesis, extracting, sorting out a high dynamic range image synthesis algorithm based on multi exposure image is adopted. By and analyzing the gray data of each exposure image, the brightness image of the light distribution of the original scene can be synthesized. By analyzing the effect of noise on the quality of the high dynamic range image, it is proposed to deal with the noise in the image before the image is synthesized. According to the characteristics of the photon shot noise, the mixed image noise problem is transformed into the average value problem of a multi exposure image sequence group, and the image visual effect is very close to the real image.