MTF Test Chart – Quality Control
I recently ordered a custom made test chart from a reputable photo lab for measuring the modulation transfer function of various imaging systems to provide unbiased, highly accurate data for future articles here on the site and on YouTube.
Bottom line upfront: The quality of the chart was not sufficient, and here is how I found out:
The Chart
The rule of thumb is: The bigger the test chart, the better the results. I therefore ordered a chart sized approximately 85x60 cm printed on Forex.
The Forex-chart.
And there is the rub – Forex is essentially fancy plastic and it has a slightly structured surface. This negatively impacts acutance of the black trapezoids, which serve as the slanted edge targets. The effect is almost not visible to the naked eye, but under the linen tester it was very obvious.
So I already knew that the chart won’t suffice, but I decided to test how big of an issue it was in practice.
The Test
For the test, I photographed the chart with an OM System OM-1 + Leica DG Summilux 25mm f/1.4 from various distances and compared the cycles per pixel-values of the same group of test patterns across the various photos.
The RAW files were processed in MakeTIFF to create a linear, relatively “untouched” TIFF.
MTF Mapper was used to calculate the following MTF values in cycles/pixel.
Distance to chart: 100cm
Distance to chart: 150cm
Distance to chart: 250cm
Evaluation
With increasing distance, the MTF values increased as well. This is indicative of poor test chart quality, and here is why:
The further the distance from the chart, the smaller the slanted edge targets in the frame. When the targets decrease in size, their fuzzy edges decrease in size as well. Therefore, the edges increase in acutance which positively affects the result of the MTF calculation.
Obviously, you want high edge acutance. If the test chart is of bad quality, you cannot resolve the issue in practice, because the distance to the chart is determined by the field of view.
Closing Remarks
Photographing a test chart at various distances and comparing the MTF results is an easy, reliable way to determine the quality of a test chart.
Note that some lenses exhibit varying performance depending on the focusing distance, so if you are unsure, test a given chart with different lenses and compare the results.
Also keep in mind that a high resolution imaging system can only be reliably tested utilizing targets of highest print quality and acutance.
Therefore, I’ve already ordered some replacement charts – this time targets printed on paper. The Forex target will serve as a fancy prop or cutting mat – I haven’t made up my mind yet. Decisions, decisions.
