Improved histogram based ghost removal in exposure fusion for high dynamic range images

For producing a single high dynamic range image (HDRI), multiple low dynamic range images (LDRIs) with different exposures are used. In high dynamic range imaging, a moving object in a set of LDRIs causes the ghost effect in a final HDRI. In this paper, we propose a ghost removal method based on histogram information. Object motion is detected by comparing ranks of intensity between LDRIs, and then the generated motion map is combined to construct the weight map in exposure fusion. Experimental results using various sets of LDRIs show that the proposed method effectively removes ghost effect and enhances contrast in a final HDRI. In terms of the computation time, the proposed method gives a better performance than existing methods.