Scene-based nonuniformity correction method using multiscale constant statistics

Abstract

In scene-based nonuniformity correction (NUC) methods for infrared focal plane array cameras, the statistical approaches have been well studied because of their lower computational complexity. However, when the assumptions imposed by statistical algorithms are violated, their performance is poor. Moreover, many of these techniques, like the global constant statistics method, usually need tens of thousands of image frames to obtain a good NUC result. In this paper, we introduce a new statistical NUC method called the multiscale constant statistics (MSCS). The MSCS statically considers that the spatial scale of the temporal constant distribution expands over time. Under the assumption that the nonuniformity is distributed in a higher spatial frequency domain, the spatial range for gain and offset estimates gradually expands to guarantee fast compensation for nonuniformity. Furthermore, an exponential window and a tolerance interval for the acquired data are introduced to capture the drift in nonuniformity and eliminate the ghosting artifacts. The strength of the proposed method lies in its simplicity, low computational complexity, and its good trade-off between convergence rate and correction precision. The NUC ability of the proposed method is demonstrated by using infrared video sequences with both synthetic and real nonuniformity. C 2011 Society of Photo-Optical Instrumentation Engineers (SPIE). IntroductionThermal array detectors, also known as infrared focal plane arrays (IRFPA), are a rapidly developing technology and are used in a wide range of industry, medical, and military applications. However, the nonuniformity in IRFPAs, which is due to pixel-to-pixel variation in the detectors' response, 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 Therefore, nonuniformity correction (NUC), being an indispensable key step, is applied to nearly all of the IRFPA-based engineering applications. Further, what makes the problem worse is that the nonuniformity varies over time and is closely related to external conditions, 2, 3 which results in the failure of traditional reference-based NUC methods. In order to solve this problem, several scene-based nonuniformity correction (SBNUC) techniques have been recently developed.There are two main categories of SBNUC: statistical methods 4-7 and registration-based methods. 8-10 Compared with registration-based methods, statistical approaches have been well studied because of their relatively lower computational complexity, smaller storage demands, and better realtime performance. The most well-known statistical method relies on the global constant statistic (GCS) assumption, 4, 5 which states that the statistics of the observed scene become constant over time. This assumption requires that each detector in the array spend an equal amount of time observing a wide range of irradiance values. So it usually needs

Scene-based noise reduction on a smart camera

“…Such methods offer good robustness but are computationally heavy and not really suitable for real-time correction. On the other hand, statistical techniques model the FPN as a random spatial noise and estimate the statistics of the noise to remove it [9]- [11]. Compared with registration-based methods, statistical approaches have been more widely studied because of their relatively lower computational complexity, smaller storage demands, and better realtime performance.…”

Scene based Non-uniformity Correction for Optical Remote Sensing Imagery

“…Such techniques provide accurate robustness however, are computationally heavy and no longer without a doubt suitable for real-time correction. On the other hand, statistical techniques version of the fixed pattern noise (FPN) as a random spatial noise and estimate the information of the noise to cast off it [22,23]. But, they depend on hypotheses concerning the sensor output image so as to separate the FPN from the actual image.…”

Regularization approach to scene-based nonuniformity correction

“…The LCS method improved the correction performance for the same number of input frames. 4 Later, Zuo et al 6 generalized the LCS method by introducing a new constraint called multiscale CS.…”

“…4 Various scene-based nonuniformity correction (SBNUC) algorithms have been proposed to solve these problems. In general, SBNUC schemes can be broadly divided into two categories: constant statistics (CS) methods [4][5][6] and least mean square (LMS) methods. [7][8][9][10] The original CS method assumes that the temporal mean and standard deviation of each pixel are constant over time and space.…”