Multi-frame linear regressive filter for the measurement of infrared pixel spatial response and MTF from sparse data

Abstract

A challenging point in the prediction of the image quality of infrared imaging systems is the evaluation of the detector modulation transfer function (MTF). In this paper, we present a linear method to get a 2D continuous MTF from sparse spectral data. Within the method, an object with a predictable sparse spatial spectrum is imaged by the focal plane array. The sparse data is then treated to return the 2D continuous MTF with the hypothesis that all the pixels have an identical spatial response. The linearity of the treatment is a key point to estimate directly the error bars of the resulting detector MTF. The test bench will be presented along with measurement tests on a 25 μm pitch InGaAs detector.

Development of a Cryogenic Test Bench for Spectral MTF Measurement on Midwave Infrared Focal Plane Arrays

“…The 2D CSIG measurement points of the transfer function are extracted by comparison between the oversampled detected interferogram and the theoretical CSIG interferogram by the use of a deconvolution algorithm using a multi-regressive filter. 7 The 2D broadband transfer function (measurement without spectral filters) first enables us to estimate qualitatively the 2D shape of the transfer function for each pixel structure. The discrete MTF measurement points are then quantitatively fitted by a model function TF f x ; f y À Á .…”

“…), 5,6 some unusual spectral effects can be expected for the future generation of infrared focal plane arrays (FPAs). Therefore, in this paper, we introduce a specific test bench for spectral MTF measurements, by the use of an interferential method using a continuously selfimaging grating (CSIG) interferometer 7 and spectral filters, in a specific cryogenic environment. The bench is tested on a particular 30 lm pixel pitch mid-wavelength infrared (MWIR) HgCdTe detector manufactured by CEA in 2011 for specific studies on low-fill-factor pixels.…”