Direct High Dynamic Range Imaging Method Using Experiential Optimal Exposure Criterion

摘要：

In this paper, we propose a simple but efficient image fusion method to combine LDR image sequence directly. The fusion results can be shown on the conventional media without tone-mapping operation. The foundation of our method is an experiential optimal exposure criterion that optimal exposure will be occurred at dramatic alteration point in LDR image sequence. In our method, each pixel value curve of LDRIs is preprocessed by chord length parameterizaiton in RGB color space. Then initial image fusion mask can be determined according to the location of maximum on the parameter curves derivative. The final HDR image can be extracted by pixel-level weighted summation using final smoothed image fusion mask. The sufficient experimentations on various scenes indicate that the proposed method can synthesize considerable HDR images for conventional media, but is more efficient than traditional HDRI methods.

关键词： component High Dynamic Range Imaging Image Fusion Chord Length Parameterization

作者: Jun Zhang Shiqiang Hu Xia Dai

作者单位: School of Aeronautics & Astronautics, Shanghai Jiao Tong University Shanghai, China School of Mathematics & Computer Engineering, Xihua University Chengdu, China

会议类型: 国际会议

会议名称: 2011 4th International Congress on Image and Signal Processing(第四届图像与信号处理国际学术会议 CISP 2011)

会议地点: 上海

会议语种:英文

页码: 760-764

在线出版日期: 2011-10-15（万方平台首次上网日期，不代表论文的发表时间）