Process and apparatus for automatically determining the modulation transfer function of focal plane array cameras

• US 20030086079A1
• Filed: 10/31/2002
• Published: 05/08/2003
• Est. Priority Date: 10/31/2001
• Status: Active Grant

• Current Assignee: Lfk-Lenkflugkã¶Rpersysteme GmbH (MBDA)
• Original Assignee: Lfk-Lenkflugkã¶Rpersysteme GmbH (MBDA)
• Inventors: Barth, Jochen, Assel, Michael, Polin, Yvonne
• Granted Patent: US 6,876,443 B2
• Time in Patent Office: Days
• US Class Current: 356/124.5
• CPC Class Codes: G01M 11/0292 of objectives by measuring the optical modulation transfer function photometry G01J
• CPC Class Codes: H04N 17/002 for television cameras

Abstract

In a process for automatically determining the modulation transfer function (MTF) of a focal plane array (FPA) cameras, a test grid is used to generate a beat whose amplitude corresponds to that of the grid image to be measured, without the sampling MTF. (The image of the test grid has a spatial frequency in the detector plane that is detuned with respect to the Nyquist frequency of the detector array. The test grid covers a range of at least half a beat period, and then, regardless of the actual phase position, the modulation amplitude is determined therefrom. Comparative MTF measurements are thereby permitted without the influence of the sampling effect of an FPA, and particularly the MTF measurement in real time suitable for the application to moving images.

Claims

1. A process for the automatically determining a modulation transfer function of focal plane array cameras, said method comprising:

providing a test grid, having an image with a spatial frequency in the detector plane that is detuned with respect to the Nyquist frequency or a rational fraction of the Nyquist frequency, of a detector array of said camera;

using said test grid to generate a beat whose amplitude correspond to amplitude of the grid image to be measured, without the sampling the modulation transfer function;

wherein

the test grid covers a range of at least half a beat period;

regardless of actual phase position, the modulation amplitude is determined therefore.

2. A process according to claim 1, wherein:

within a line or column, the gray-value waveform of the image of the grid is read out; and

thereafter, minimum and maximum gray values are determined, the difference between the minimal and maximal value corresponding to transmitted contrast.

3. The process according to claim 1, wherein;

from measured gray values, a mean value is determined and is subtracted from all measured values to form converted values;

thereafter, the converted values are sorted corresponding to

$n=0,2,4,\mathrm{\dots <br><br>}=2<br><br>v;<br><br>\text{\hspace{1em}<br><br>}<br><br>f<br><br>\left(x_{2<br><br>v}\right)={\left(-1\right)}^v<br><br>\sin <br><br>\left(\frac{2<br><br>\mathrm{\pi <br><br>}}{P^{*}}*2<br><br>v\right)$$\mathrm{and}$$n=1,3,5,\mathrm{\dots <br><br>}=2<br><br>v+1;<br><br>f(x_{2<br><br>v+1)}={\left(-1\right)}^v<br><br>\cos <br><br>\left(\frac{2<br><br>\mathrm{\pi <br><br>}}{P^{*}}*\left(2<br><br>v+1\right)\right)$

whereby in each case half of all measured values are available for a sine fit and a cosine fit, in each group a preceding sign of every second measured value being inverted.

4. The process according to claim 1, wherein:

amplitude of the beats is determined directly, without a fit, by subtracting a mean value from the measured gray values;

the sum of squared of the gray values of adjacent pixels respectively results in a square of the modulation amplitude; and

the thus obtained amplitude values are averaged over half a period length or an integral multiple thereof.

5. Apparatus for implementing the process according to claim 1, comprising:

an energy source;

a test grid having a grid structure for masking energy from said source to form a pattern;

means for focusing said pattern on a detector array in a focal plane camera; and

an image processor couple to receive an output signal from said camera;

wherein the test grid has spacer markers whose spacing corresponds to length of the grid structure.