Read Light intensity from a ROI

[Rlima]

    Hi All...

     I'm currently working on a proyect wich I need to test test  light intensity from a radio faceplate, so far i have

develop some testing using the light Intensity vi.  I'm getting data (values from 0 to 255) I'm checking diferent ROI's

(all buttons for uneven ilumination) in the same picture,  I need to define a set of limits (values from a master face plate) 

or we can called calibration.   so I can judge production faceplates.  I have many variations in the data from one unit to other is not stable

     I'm using a  15 megapixes camera  (Gigabit) in a dart enclouser to take the picture..

      I'm having a hard time to figure out how to make it work.

      I'll really apreciate you help,

      See the pictures  attach

          #1  UUT-Photo Master 

         #2  UUT-Photo Key4 NG           production unit

         Best Regards..

          Ing. Reynaldo Lima V.

          INGEL (Ingenieria Electronica)

          Email:  reynaldolima@aol.com

[muks]

I have seen the images. There is an extra white dot in the image what is that? is that the difference you need to find? why can you manually select different roi and find their histogram distribution?

[Jörn]

Maybe the intensity light meter VI are not optimal. Maybe you should extract the items using thresholding the image. Then convert the thresholded image to a set of ROI , use these ROI (which would enclose only the bright areas) in the manner as in the intensity light meter VI. Then you should get a good measure of the intensity of the items.

Which of the intensity light meter VIs are you using?

If you pick only one point the Values can vary according to noise.

If you use the rectangle you will aquire many background pixels and the mean calculated intensity depends on the amount of background pixels.

Message Edited by Jörn on 2009-11-11 11.11.2009 10:32 PM

[jhc2]

Reynaldo

The easiest way to get some measure of reflectance, which I think is what you're trying to measure, from a number of different samples is to include a known reference in the camera's field of view. This assumes that the illumination doesn't vary and is even across the area being imaged.

You can do this with a small piece of  spectralon. With constant illumination, the spectralon will be an area of known reflectance in each image. Then you can compare the reflected values from your face plates to the reflected value from the spectralon. It will act as a built-in calibration region.

You can buy spectralon pieces in almost any shape or size.