A Voxel-Based Adaptive Exposure Model for Physically Based Volume Rendering

In physically based volume rendering of volumetric medical imaging data, exposure levels are typically set empirically. However, differences in voxel values and opacity across transfer functions can make a fixed setting produce underexposed or overexposed results. To address this issue, we propose a voxel-contribution-based adaptive exposure model for physically based volume rendering. During volume integration, the method visibility-weights effective volumetric-sample energies using accumulated transmittance and opacity and estimates a global exposure coefficient from their weighted geometric mean. Compared with a two-dimensional HDR method operating on the rendered image, the proposed method incorporates three-dimensional information from volume integration into exposure estimation. The mean exposure-coefficient deviation is approximately $0.005\%$, compared with $0.031\%$ for the two-dimensional method, while output mean-luminance variation remains below $1.10\%$ over the tested range of normal illumination. The results demonstrate stable exposure and structural response across the evaluated transfer functions and illumination conditions.