2

Calculating EV and then Lx is pretty simple. The problem starts when trying to calculate these by using EXIF data that come from different cameras.

I tried capturing a static object in the exact same position, angle and light condition, with 2 different cameras, and the EV result was completely different.

The thing is, I’m not sure how to treat the various parameters which I believe have dramatic influence on the light value, e.g. focal length. In all equations that I found on the internet, the only parameters specified were Aperture, Exposure and ISO.

Does anyone know how to handle this situation?

mattdm
  • 143,140
  • 52
  • 417
  • 741
SMD
  • 123
  • 1
  • 3

2 Answers2

1

I will try my best to answer this. I will first explain that there is a difference between LV, EV and Lux.

LV: Light Value is a simplified figure of Lux and is an industry standard. Please refer to the table below:

LUX     LV
10240   12
5120    11
2560    10
1280    9
640     8
320     7
160     6
80      5
40      4
20      3
10      2
5       1
2.5     0

EDIT: It seems I had this very wrong. I have updated it.

Camera manufactures generally use a light box at the value of 12 for adjustments. So, the light value for 12 is 12,000 Lux.

Lux: Lux (as quoted above) is a unit of measuring light, or luminance. Please refer to the table below:

Examples 

Illuminance        Surfaces illuminated by:
0.0001 lux         Moonless, overcast night sky (starlight)[3]
0.002 lux          Moonless clear night sky with airglow[3]
0.27–1.0 lux       Full moon on a clear night[3][4]
3.4 lux            Dark limit of civil twilight under a clear sky[5]
50 lux             Family living room lights (Australia, 1998)[6]
80 lux             Office building hallway/toilet lighting[7][8]
100 lux            Very dark overcast day[3]
320–500 lux        Office lighting[9][10][11]
400 lux            Sunrise or sunset on a clear day.
1000 lux           Overcast day;[3] typical TV studio lighting
10000–25000 lux    Full daylight (not direct sun)[3]
32000–130000 lux   Direct sunlight

EV:

Exposure Value is really Exposure compensation. If you change this value, you are not changing the Lux being shone into the camera. The only way to change the Lux is to change the lighting.

Above, you saw before that the Light Value and Lux are related. Now I'll go into more detail of the LV.

For a camera to be exposing correctly, it is said that it must have the following consistencies:

   LV       = 12 (or 10,240 Lux)
   ISO      = 100
   Aperture = f5.6
   Shutter  = 1/125

For repairing and adjusting cameras, the light boxes are valued at LV and temperature. LV is just an industry standard to simplify the Lux value. If you want exact measurement, a lux meter would be your best bet.

While the above table is also a standard, it may vary camera to camera (probably not more than a stop. I haven't done enough testing, though). If you're trying to calculate the LV/Lux, it'll be best to look at the Lux table above.

BBking
  • 2,765
  • 1
  • 16
  • 27
0

In fact this is a difficult topic with many misconceptions I'll try to say what I think however have a look to this white paper. Sincerely I would not speak of luminance of different cameras because luminance is:

a photometric measure of the luminous intensity per unit area of light travelling in a given direction. It describes the amount of light that passes through or is emitted from a particular area, and falls within a given solid angle. [Wikipedia]

I rather say light sensitivity of different cameras. It is easy to compare relative performance of different cameras taking a photo of the same target with the same distance and same light you can take a ROI and measure the mean gray level (e.g. with ImageJ Ctrl+M). Of course you have to face with the image quality varying with different factors (the first: ISO) so you should express the performance with a ratio (e.g. garylevel/noise or graylevel/integral(MTF)). You can take it to a radiometric level using lamps with a certify lux and reflectance standards (PTFE, Barium Sulphate, white paper all this materials have reflectance near 99%). So now you can calculate the dependencies of all these "Exif" parameters and find the law that describe them. After this you can make the next step to the solution and find the contribute of all the others "non-Exif" parameters such sensor size, sensor sensitivity, lens quality, lens dimension so definitively light-gathering power! It's nearly impossible to calculate separate the interaction between all theese factors but with a bit of math you can find the equation that fit best the results you obtain not knowing "how it came out". However it's impossible to calculate it only from the exif file because exif file doesn't describe all the aspects of your optical system.

It's over? Maybe not because the sensitivity of your sensor probably is not linear, for better result you should make all this stuff again with more or less light.

G M
  • 413
  • 2
  • 15