Abstract
Proposed herein is a motion-adaptive automatic exposure (AE) system that dynamically reshapes an image signal processor (ISP) weighting table using aggregate motion heat maps. Motion data accumulated over a configurable temporal window identifies high-activity regions, such as roadways or crosswalks, and converts those regions into higher exposure weights, while static regions such as sky or water receive lower weights. A minimum baseline weight remains in low-activity regions to preserve forensic visibility if an unexpected event occurs. Unlike center-weighted, spot, matrix, or object-detection-based metering, the system learns where activity occurs without requiring classification of a person, vehicle, package, or other object. A software-defined interface supplies the resulting coefficients to the ISP, which applies the heat-map values as multipliers or additive offsets to an existing weighting table. The hardware-agnostic approach supports security and surveillance cameras across ISP architectures while reducing manual calibration and computational load.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
McCauley, Adam Grant, "DYNAMIC EXPOSURE WEIGHTING BASED ON MOTION HEAT MAP ANALYSIS", Technical Disclosure Commons, (September 24, 2026)
https://www.tdcommons.org/dpubs_series/11849