You can do the FFT of the interference pattern. That will give you the reflectivity profile of your sample from where you can measure the thickness.

Thanks a lot. It will be great if you could some more details like thy type of light source, and the basic image processing.

We are using a swept-source laser as the source of an optical coherence tomography, which can be used to measure film thickness, but probably there are simpler (and less expensive solutions out there).

The processing is quite straightforward in MATLAB:

% Do the FFT of your scan

% Bmodescan is your interference pattern, nSamplesFFT is the number of samples you want for your FFT (a power of two, to speed things up)

fftBscan = fftshift(fft(double(Bmodescan), nSamplesFFT,1),1);

% Keep only the single-sided FFT (left half of the spectrum)

fftBscan = fftBscan(end/2:-1:1,:);

% Compute the modulus of the FFT

struct_image = abs(fftBscan);

Thanks a lot. I have limited resources at the moment. Using a He-Ne laser, I am able to produce fringes and record it using a CCD. I am able to view the variation of fringe spacing according to the thickness variation, but not clear how to extract thickness value out of it. I do know the refractive index of the film.

Ok, I see. I would extract a line of the image from the center of the fringe pattern and then apply the processing presented above, more or less what they do in http://www.hamamatsu.com/jp/en/technology/innovation/spectroscopic/index.html

I think you will need also the spectral width of your light source to properly calibrate the horizontal axis of your fft.

I know for this measure it is polarimetry method that is commonly used. The full process should be quite simple but there are several stages and I wouldnt like to go into all details, just google for it (also Im not a big expert on this). It can yield data on both thickness as well as index, and each additional layer requires more measures. Generally you launch a linearly polarized light at certain incident angle, get whatever S.O.P that is reflected from the 1st surface and the one beneath. then, using polarization converter / controller you reverse the signal back to a linearly polarized light. The observer should then zero the amount of light in the eyepiece / camera, such that the reflected signal is orthogonal with a linear polarization filter. Then a numerical computation is run and the result is found graphically on a certain abstract map.

You can actually do this by setting up a basic Michaelson interferometer with a white lamp. The short coherence length of white light gives you a very sharp response on interfaces. If you can get enough reflections for the front and back surfaces of the thin film, you will see interference only when the reference arm is at the same displacement as the surface. By observing the change in reference displacement as you go from one interference pattern (front surface) to the other (back surface) you effectively measure the optical path length of the thin film.

What is the thickness range of your oil film on the glass substrate. For films less than about 100 microns you can use spectral reflectance using a broad band width light source, and for films from 10 microns and up you can use Low coherence interferometry with instruments such as the Lumetrics Optigauge.(see http://www.lumetrics.com/technology.html)

First of all u have to record the double exposure hologram of substrate. Then the interference fringes will be revealed on it. By counting the total no of fringes and measuring the fringe width it is possible to measure the thickness of of film on glass substrate

Oil has a very similar refractive index to the glass substrate. It is not clear your setup but If you observe interference fringes ask yourself if they are not comming from the substrate sides interaction.

Suggestion: Under focused light observe reflected light. (you need to expand the oberved beam and reduce the exploration área to reduce thickness non-uniformity within the beam). Before that, elliminate substrate back reflection with a black second substrate (black anodized aluminium e.g.) index matched to the glas substrate. Another oil or glycerine could be ok for index matching. Alternatively if the backside of the glass is rough even better (then you elliminate completely the second reflection). If you still distinguish fringes then you have a measurable situation. If not the case then forget the use of the laser....

Hi there !

With a single interferogram you can not determine without ambiguity the thickness of the oil film. In interferometric metrology the phase-shift interferometry with monochromatic light source is commonly used (see link attached). Otherwise, I concur with Craig above, you could use with light interferometry in combination with a phase-shift device or, even simpler, if your oil film is thin enough, you could "just" look at the interference color created when illuminating the oil film with an spectrally incoherent light source (as in soap bubbles or oil spilled in the street).

http://fp.optics.arizona.edu/jcwyant/Short_Courses/SPIE-7-30-2000/Session%209.pdf

I forgot to mention : you actually determine the product of the thickness and the refractive index. The last is known from you apparently.