Scene-based nonuniformity correction (NUC) was performed via two techniques, and the results are presented in terms of ability to allow increased sensitivity and to minimize scene degradation. The two techniques perform recalibration in real-time based on the radiance levels of the scene being viewed. The scene-based NUC was effected with an algorithm based on a temporal high-pass filter, and with one based on an artificial neural network. Recorded data from an MWIR staring array collecting different images were used for the experiments. The advantages of the two scene-based NUC techniques are summarized and compared to those of the traditional calibration technique. The potential for the elimination of spatial noise and the achievement of BLIP performance levels in high-quantum-efficiency FPAs are emphasized. Spatial noise can be eliminated by employing real-time signal processors.© (1990) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.