HDR imaging based on region segmentation

In high dynamic range(HDR) imaging, artifacts, generated on the block boundaries, has been a puzzling problem all the time. A novel HDR imaging method is proposed to eliminate the fatal defect in this paper. Firstly, we segmented the low-pass filtered images using the region segmentation method. Secondly, the most informative one in corresponding regions is selected and all the selected regions are mosaiced. Finally, a weighted function is used to eliminating the effect of the artifacts. The experimental results show that the proposed algorithm can greatly enlarge the dynamic range without losing the details of images, which has a wide application prospect in both civil and military fields.

[1]  Kang-Hyun Jo,et al.  HDR Image Generation based on Intensity Clustering and Local Feature Analysis , 2011, Comput. Hum. Behav..

[2]  Tang Yinggan.  Image Segmentation Based on Two-Dimensional Bound Histogram Fisher Criterion and Its Fast Recursive Algorithm , 2009 .

[3]  K. R. Ramakrishnan,et al.  High Dynamic Range Imaging , 2013, 2013 International Conference on Communication Systems and Network Technologies.

[4]  Y. Shimodaira,et al.  Gradient Based Synthesized Multiple Exposure Time HDR Image , 2007, 2007 IEEE Instrumentation & Measurement Technology Conference IMTC 2007.

[5]  K. Jo,et al.  Recursive HDR image generation from differently exposed images based on local image properties , 2008, 2008 International Conference on Control, Automation and Systems.

[6]  Annamária R. Várkonyi-Kóczy,et al.  Fuzzy based brightness compensation for high dynamic range images , 2005, 2005 IEEE International Conference on Intelligent Engineering Systems, 2005. INES '05..

[7]  E. Reinhard.  Photographic Tone Reproduction for Digital Images , 2002 .

[8]  A. Ardeshir Goshtasby,et al.  Fusion of multi-exposure images , 2005, Image Vis. Comput..

[9]  Jitendra Malik,et al.  Recovering high dynamic range radiance maps from photographs , 1997, SIGGRAPH.