TECHNOLOGY AREA(S): ElectronicsOBJECTIVE: The objective of this project is to develop a novel technique to perform Non-Uniformity Corrections (NUC) of infrared focal plane arrays by minimizing and potentially eliminating the need to use active thermal reference sources. The desired approach is to use an in Dewar variable aperture or other similar mechanism to generate the gain and offset parameters in place of active thermal reference sources and thereby eliminate the resultant loss of live infrared imagery to correct staring arrays. The results of this project will be used by the Government to assess the feasibility to eliminate the cost of the active thermal reference sources and reduce the time required to perform the NUC.DESCRIPTION: Sensors using cooled staring focal plane detector arrays will typically need some form of NUC to suppress random noise such that the scene can be viewed accurately. The pixel-to pixel variations are typically characterized as differences in gain and offset. The gain parameter refers to the slope of the pixel output (rate of change) response versus the input signal and the offset parameter is a fixed additive value unique to that specific pixel. The non-uniformity issue arises when each neighboring pixel in the focal plane array has a different gain and offset which leads to a fixed pattern noise which can dominate the output video unless it is electronically compensated. A standard method of compensation involves blocking the focal plane array with a known uniform source at different temperatures to measure each pixel’s gain and offset values and then apply these offsets to each pixel on live output video to remove the fixed pattern noise. The downside of this approach is that while the thermal reference source is being used for a NUC the focal plane array is blocked from viewing the outside scene and thereby blinding the operator.The 3GEN FLIR is a new technology for combat vehicles that incorporates a variable aperture mechanism (VAM) inside the cold space of the Dewar to vary the F# to enable the sensor to switch long focal length optics without increasing the aperture size. The aperture mechanism is structured like a conventional iris such as used in photographic camera optics. This means that the change in aperture shape will vary the amount of light flux from a given object source in a controlled and known fashion. By including the VAM in the NUC process to change the incoming photon flux instead of the thermal reference source, the NUC could be performed faster and not block the live image from the sensor operator. This new technique could be applied to a full two point correction calibration at system turn on and one point offset correction during live scene viewing thus eliminating blackout time of the sensor. This Novel NUC approach would eliminate blind time, which is extremely important to the warfighter that is common with the conventional NUC approach that uses thermal reference sources.PHASE I: Develop the non-uniformity process sequences, algorithm, and performance metrics for one and two point simulated corrections using an in Dewar variable aperture mechanism or similar mechanism in the optical chain without the presence of an active thermal reference source compared to the conventional approach with a thermal reference source. Use Mat Lab or equivalent software to show with at least static imagery that the variable F# technique (either in Dewar or in the warm space) can correct the spatial noise of a longwave focal plane array to less than 25% of the temporal noise. Also, demonstrate the compatibility of the process algorithm with other scene-base non-uniformity correction methods that may be running in parallel. Deliver documentation of the work effort and the results in a technical report per DI-MISC-80048.PHASE II: Using a 3GEN Dewar (this will be a GFE), produce a breadboard apparatus and perform laboratory and field tests in both static and on the move scenarios with live video-rate processing to obtain non-uniformity corrections that result in the spatial noise to be less than 25% of temporal noise. Measure performance over an eight hour window and check the stability of the corrections. Document and deliver the optimum processing algorithm, actual test apparatus hardware details, and the test raw data, analysis, and conclusions per DI-NDTI-80809B.PHASE III: The end state is the 3GEN FLIR which will eliminate the Thermal Reference Source and use the variable aperture for the non-uniformity correction. The transition will be Low Rate Initial Production (LRIP) and full rate Production of the 3GEN FLIR through an engineering change proposal. Having Non-uniformity correction capability that doesn’t block the visual scene can be applied across DOD and industry for Night Vision Devices.KEYWORDS: Variable Aperture Mechanism, 3GEN FLIR, Non uniformity correction, infrared, offset, gain, focal plane arrayReferences:G. Bieszczad, et. Al., “Method of detectors offset correction in thermovision camera with uncooled microbolometric focal plane array.” Proc. SPIE Vol. 7481, 2009; P. Fillon, A. Combette, P. Tribolet, “Cooled IR detectors calibration analysis and optimization,” In Proc. SPIE Orlando, 2005, [5784-42]; A. Kumar, S. Sarkar, and R.P. Agarwal, “Fixed pattern noise correction and implementation for infrared focal plane array based staring system using scene statistics,” International Journal of Imaging Science and Engineering (IJISE), vol. 1, no. 1, January 2007 [ISSN:19349955]; K.G. Lesueur, E. Jovanov, and A. Milenkovic, “Lookup table based real-time non-uniformity correct o infrared scene projectors,” Proc. Of the 12th Annual DoD High Performance Computing Modernization User Group Conference, Austin, TX, June 2002.; Computational Sensors Corp “CSC-341-IR ROIC with analog Domain Bad Pixel and Nonuniformity Correction.”; US Patent #9,602,744 “Method of Detector Gain and Offset Level Estimation by Means of a Varaible Aperture Transmission Mechanism”, Hall & Bourgeois, 21 March 2017.
Show All
