Target-oriented adaptive brightness algorithm for EBAPS imaging systems

The key distinction between Electron Bombardment Active Pixel Sensors (EBAPS) and conventional CMOS lies in their external photoelectric effect-driven photocathode, which converts photons to electrons. Applying high voltage (HV) accelerates these photoelectrons to bombard the CMOS surface, triggering secondary electron emission for intrinsic charge multiplication. During nighttime operation, EBAPS must autonomously regulate photocathode HV to maintain display consistency across varying illumination by controlling electron bombardment gain dynamically.This study addresses EBAPS optimization under photometric feedback-free conditions by proposing a target-feature-driven adaptive brightness control algorithm. The algorithm extracts spatio-temporal brightness distributions of target regions in real time, establishes feature-to-illuminance mapping, and intelligently adjusts HV parameters for self-adaptive imaging. Experimental results demonstrate: 1) Within 5×10⁻⁴ lx to 1 × 10⁻² lx ultra-low illuminance, brightness fluctuations are stabilized within ±5%; 2) Under dynamic target displacement and morphological variations, system brightness error remains below ±3%. By synergistically analyzing spatial distributions and temporal correlations of target features, the algorithm enables closed-loop brightness control while accommodating small moving targets.