Kalman filtering for adaptive nonuniformity correction in infrared focal-plane arrays

Abstract

A novel statistical approach is undertaken for the adaptive estimation of the gain and bias nonuniformity in infrared focal-plane array sensors from scene data. The gain and the bias of each detector are regarded as random state variables modeled by a discrete-time Gauss-Markov process. The proposed Gauss-Markov framework provides a mechanism for capturing the slow and random drift in the fixed-pattern noise as the operational conditions of the sensor vary in time. With a temporal stochastic model for each detector's gain and bias at hand, a Kalman filter is derived that uses scene data, comprising the detector's readout values sampled over a short period of time, to optimally update the detector's gain and bias estimates as these parameters drift. The proposed technique relies on a certain spatiotemporal diversity condition in the data, which is satisfied when all detectors see approximately the same range of temperatures within the periods between successive estimation epochs. The performance of the proposed technique is thoroughly studied, and its utility in mitigating fixed-pattern noise is demonstrated with both real infrared and simulated imagery.

Extended Recursive Filtering Estimation of Detector Offset Nonuniformity in Infrared Imaging Systems

“…the full dynamical range of the camera. Then the α and l parameters can be obtained from [5]. If a short block of frames is employed, i.e.…”

“…Our group has been active in the development of novel scene-based algorithms for NUC based on statistical estimation theory. In previous works, we have developed a Gauss-Markov model to capture the slow variation in the FPN and such model was used to adaptively estimate the NU within blocks of IR video sequences using different types of Kalman filters [5,6,7].…”

“…In this paper, the aforementioned recursive filter is extended adding to the intra-block processing, where the offset is assumed to be constant, an inter-block processing algorithm that considers the slow and random temporal drift in the offset NU. The extension to the algorithm relies on the following assumptions: i) the input irradiance is a random white process following an uniform distribution in a range that is common to all detectors; ii) within each block of frames the offset NU is assumed to be constant; and iii) the slow and random drift is modeled by a discrete-time Gauss-Makov random process [5,6,7,8]. The extended filter is designed to continuously compensate for the NU when new blocks of frames arrive, and more importantly, the assumption on the stationarity of the offset NU is relaxed.…”

Acceleration algorithm for constant-statistics method applied to the nonuniformity correction of infrared sequences

“…It is well known that nonuniformity noise in IR imaging sensors, which is due to pixel−to−pixel variation in the detec− tor's responses, can considerably degrade the quality of IR images since it results in a fixed−pattern−noise (FPN) that is superimposed on the true image [1]. Currently there are two main fields in correcting the noise affecting such images (technically called non−uniformity correction): one corre− sponding to the reference−based approach, which uses inter− polation techniques; and the second to the correction based on the scene, which requires parameter estimation algo− rithms to correct image online.…”

Minimum Variance Gain Nonuniformity Estimation in Infrared Focal Plane Array Sensors

“…In addition, what is worse is that the NU varies slowly on time, depending on the type of technology that is been used. In order to solve this problem, several scene-based NUC techniques have been developed [3,4,5,6,7,8,9]. Scene-based techniques perform the NUC using only the video sequences that are being imaged and not requiring any kind of laboratory calibration technique.…”