How can sensor dimensions be estimated from pixel size, megapixels, and aspect ratio?

Asked 4/10/2016

1 views

2 answers

0

If you know a camera’s total pixel count, aspect ratio, and approximate pixel pitch (for example, 1.5 µm), how can you estimate the physical sensor dimensions and area? In the example discussed, the sensor was said to be about 15% larger, with 8 MP and a 4:3 aspect ratio. I understand that you could estimate area from the percentage increase and aspect ratio alone, but how does pixel size fit into the calculation?

Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0

Photography Stack Exchange contributor

10y ago

2 Answers

3

They are two independent sums which give (roughly) the same number:

  • 15% bigger sensor: 15.5 mm2 * 1.15 = 17.8 mm2.
  • 1.5 micron pixels: knowing that the iPhone 5S has an 8 MP camera and a 4:3 aspect ratio means you know that it's a 3.3 x 103 by 2.4 x 103 array. Multiplying those numbers by 1.5 microns gives you 18 mm2.

None of that is rocket science, but remember that comment was posted on the iPhone 5S launch day when you couldn't just go and look this stuff up on the web. The fact that both numbers are in the same ballpark means that it was likely that the calculation was approximately right.

Originally by user11371. Source · Licensed CC BY-SA 4.0

user11371

10y ago

0

AI Answer

Generated from our catalog & community — verify before relying on it.

Pixel pitch gives you another way to estimate sensor size.

For an 8 MP sensor with a 4:3 aspect ratio, the pixel dimensions are approximately 3264 × 2448 pixels. If each pixel is about 1.5 µm wide, then the sensor dimensions are roughly:

  • width: 3264 × 1.5 µm ≈ 4.9 mm
  • height: 2448 × 1.5 µm ≈ 3.7 mm

That gives an area of about 18 mm².

The other estimate starts from the previous sensor area and the claimed increase. If the old sensor was about 15.5 mm², then a 15% increase gives:

15.5 × 1.15 ≈ 17.8 mm²

Those two independent estimates are close, which suggests the result is reasonable.

So yes: if you already know the new area and the aspect ratio, you can derive width and height from those alone. Pixel size is useful because it provides a separate cross-check using resolution × pixel pitch.

It’s only approximate, though, because real sensors are not perfectly filled with active pixel area; there can be gaps and inactive regions. More accurate methods use published crop factor/specifications or field-of-view measurements.

UniqueBot

AI

10y ago

Your Answer