重新思考自动曝光控制:一种具有语义引导的物理感知多流框架OA

Auto-exposure(AE)is a pivotal component in imaging systems,playing a decisive role in achieving bal-anced image brightness and enhancing the accuracy of high-level vision tasks.However,existing techniques face consid-erable challenges:traditional rule-based algorithms are constrained by the"semantic gap"and struggle with semantic ambiguities in complex lighting conditions,while end-to-end deep learning approaches frequently operate as physically unconstrained"black boxes"leading to significant temporal instability.To address these issues,this paper introduces a physics-aware white-box auto-exposure framework,named PhysAEC.Departing from traditional parameter regression and image enhancement methods,we redefined the core AE challenge as a"multi-target luma prediction"task to estab-lish optimal exposure anchors for the ISP control loop and ensure semantic adaptability and physical interpretability.Phy-sAEC adopts a three-stream decoupled architecture to facilitate the integration of heterogeneous information:an RGB se-mantic stream extracts high-level scene priors to eliminate semantic ambiguities(e.g.,in backlight scenarios),while the raw-domain spatial grid and global histogram streams provide precise local intensity distributions and dynamic-range boundary constraints,respectively.Furthermore,to mitigate temporal oscillation during continuous inference,we intro-duced a tolerance-aware loss(TAL)that incorporates the hysteresis characteristics of photometric control.By optimiz-ing physical regularization at the target level,TAL effectively suppressed parameter jitter resulting from minor fluctua-tions.Experiments conducted on our Balanced-AE-Dataset,comprising 10000 high-quality samples,revealed that Phy-sAEC achieves a prediction accuracy of 94.05%under standard conditions,with the mean absolute error decreasing from 21.12 to 2.53.In complex high-dynamic-range scenarios,the method yielded a PSNR of 38.98 dB and an SSIM of 0.994.These results underscored the proposed method's successful integration of semantic understanding and robust physical control,establishing a new paradigm for low-level ISP control tasks.