Why does my MTF calculation from a printed bar chart just reproduce contrast, and how should I normalize it?
Asked 6/7/2021
4 views
2 answers
0
I'm trying to calculate an MTF curve from photos of a printed black-and-white bar chart. My current approach is to scan image columns, find luminance min/max values, compute contrast with C(f) = (Vmax - Vmin) / (Vmax + Vmin), then apply MTF(f) = 100% × C(f) / C(0). But the result is basically the same as the contrast curve, and I'm not sure how to obtain a properly normalized MTF or a single sharpness score. I'm also using a matte-paper printout and a smartphone photo of the chart. Is the issue my formula, my chart, or both?
Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0
Photography Stack Exchange contributor
5y ago
2 Answers
1
From my previous comment to your previous question:
Horizontal and vertical bars do not, strictly speaking, measure tangential and sagittal performance. Sagittal lines are like the spokes of a wheel spreading out from the center of the lens' field of view. Tangential lines are tangent to the circular surface of the wheel that rolls over the ground.
Further, matte paper is notorious for blurring sharp edges. You're starting with a poor quality chart. All that follows will suffer the consequences for that decision. There's a reason real test charts are expensive. They're held to very high standards of quality control and use more expensive printing methods to achieve them.
Originally by user15871. Source · Licensed CC BY-SA 4.0
user15871
5y ago
0
Generated from our catalog & community — verify before relying on it.
The main issue is your test target, not just the math. A bar/grid chart does not provide the low-frequency reference needed for proper normalization, so C(0) is not meaningfully defined. In that case, applying MTF(f) = C(f)/C(0) can just reproduce the same curve shape.
Also, a matte-paper printout is a poor sharpness target because the paper and printing process blur edges before the camera even sees them. That limits the validity of any MTF result.
A better approach is to use a proper optical test chart, especially a slanted-edge target, which includes low-frequency information and is the standard way to derive MTF in tools like Imatest.
So, based on the answers:
- your printed bar chart is not suitable for accurate MTF normalization
- matte paper introduces blur and reduces chart quality
- horizontal/vertical bars also don’t directly represent sagittal/tangential lens behavior
If you want a meaningful sharpness ranking, use a proper slanted-edge chart and derive MTF from that. Any single “average sharpness score” from your current setup would be unreliable.
Recommended products
UniqueBot
AI5y ago
Your Answer
Related Questions
Can an MTF chart tell you anything about a lens’s bokeh quality?
Should I adjust Lightroom's Basic panel or Tone Curve first when setting contrast?
Can a slanted-edge SFR/MTF test measure lens resolution beyond the sensor’s Nyquist limit?
How can I make product photos brighter and more contrasty without changing the colors too much?
How can I get more contrast in a darkroom print when I'm already using a grade 5 filter?