Random transparency targets for modulation transfer function measurement in the visible and infrared regions

Abstract

A major problem encountered in measurement of MTF in sampled-image systems (such as staring or scanning imagers) is that the image modulation depth depends on the position of the deterministic target (typically a point source, line source, or bar target) with respect to the image sampling locations. 1-3 A random transparency test target of known spatial-frequency content allows measurement of a shiftinvariant MTF because the information of the test target has random position with respect to the sampling sites. Random test targets have been used previously for MTF measurements of nonsampled systems such as camera lenses4 and telescopes.5 An alternative technique, the scanning-knife-edge method,6 has been developed to address the issue of shift variance in MTF measurement of imager systems that use focal-plane arrays (FPAs). A tilted knife edge is scanned across a column of the FPA. Our random-target method does not require mechanical scanning and tests the entire image plane at one time. The random-target method creates random test patterns similar to those created using the laser-speckle method,7 but on a broadband basis, without the use of laser radiation. This allows use of blackbody radiation for the source. This method tests both the FPA and the optics together, whereas the speckle MTF tests only the FPA.

Measurement of the point-spread function of a noisy imaging system

“…When a focal plane array (FPA) images the grating the discretization means that the FPA must be moved relative to the image to give minimum and maximum MTF curves [17]. Sinusoidal laser interference patterns have been used by Marchywka and Socker [18], and laser speckle techniques can also be employed [19]. The main limitation of lasers is the monochromatic light.…”

Investigation of a CCD modulation transfer function using the speckle method at different laser wavelengths and sub-windowing options

“…The monochromatic laser speckle pattern is widely used in stochastic MTF measurement methods, where a target with random distributed intensities from dark to the "white" patterns with a known spatial content is recorded by the imaging sensor [13,21]. The output power spectral density (PSD) of the speckle pattern recorded by a CCD sensor is given by [16]:…”

“…The idea of using targets with random distribution of Article published by EDP Sciences International Journal of Metrology and Quality Engineering the spatial content firstly was suggested in [9], where it was shown that the spatial frequency distribution of the photographic emulsion silver grains on the uniformly exposed chart pattern is ergodic, and its autocorrelation function can be described analytically. This method has become widespread and is widely used along with a test input pattern generated by the laser speckle illumination [10][11][12][13][14][15][16][17]. The laser speckle method has several main advantages compared to other existing methods: (a) it does not require any imaging optics, (b) the alignment of the optical system is not critical as with the bar chart test inputs, (c) it allows characterizing entire area of the sensor.…”

Subpixel Point Spread Function Estimation from Two Photographs at Different Distances

“…Several patterns have been used for psf estimation, ranging from pin-hole, slanted-edge [18,30,38,11], or arc-step-edge patterns [20,19] to random noise images [12,21,2,3,7]. Until recently, even non-blind subpixel psf estimation methods reported in the literature led to ill-posed inverse problems.…”