Multi-Exposure Fusion with Adaptive Well-Exposedness Weight Factor and Enhanced Guided Filtering

A multi-exposure fusion method is presented to optimize weight factor of well-exposure and the global weight map for multi-scale image fusion. In the method, the weight factors of local contrast weight factors, well-exposedness and color inconsistency are considered to estimate initial global weight map, where the well-exposure weight factor is adaptively adjusted for the intensity of the exposure, and the details of the over-exposure and under-exposure are perfectly preserved. Additionally, the enhanced guided filtering with adaptive adjustment of regularization parameters according to the gradient of the image is employed to de-noise and maintain continuity. Experiment results demonstrate that the weight detail optimization method achieves a certain gain for existing multi-scale and multi-exposure image fusions, which provides a feasible optimization scheme for multi-exposure fusions for consumer-level cameras.