In this study, we have created a bridge, which can connect the reference-based NUC and scene-based NUC. The right probability of the scene-based NUC parameters was calculated based on the Bayesian framework. The right probability composed of prior and observation probability was used to determine whether the calculated scene-based NUC parameters are suitable to correct the nonuniformity. The local same distribution constraint is defined in this paper, and the Infrared Focal Plane (IRFPA) gain space relativity has been discovered from the reference-based parameters by this paper firstly. The Bayesian prior probability is mainly determined by the local same distribution constraint, and the Bayesian observation probability is mainly determined by the Infrared Focal Plane (IRFPA) gain space relativity. This method can effectively balance the relationship between convergence speed and ghosting artifacts. Finally, the real and simulated infrared image sequences have been applied to demonstrate our algorithm's positive effect.In this study, we have created a bridge, which can connect the reference-based NUC and scene-based NUC. The right probability of the scene-based NUC parameters was calculated based on the Bayesian framework. The right probability composed of prior and observation probability was used to determine whether the calculated scene-based NUC parameters are suitable to correct the nonuniformity. The local same distribution constraint is defined in this paper, and the Infrared Focal Plane (IRFPA) gain space relativity has been discovered from the reference-based parameters by this paper firstly. The Bayesian prior probability is mainly determined by the local same distribution constraint, and the Bayesian observation probability is mainly determined by the IRFPA gain space relativity. This method can effectively balance the relationship between convergence speed and ghosting artifacts. Finally, the real and simulated infrared image sequences have been applied to demonstrate our algorithm’s positive effect.