Have you considered computerized tomography (CT)? In general, signals are transmit through an object at different angles. Properties of the signal that are affected by the medium through which it has passed (e.g. attenuation, phase delay, Faraday rotation) are measured by the receiver to form projections of the object interior. The projection data and transceiver geometry is used to reconstruct the interior of the object (potentially the interface between two objects) in a non-invasive manner.

See Principles of Computerized Tomographic Imaging by  Kak, A.C. and Slaney,M.(1998) for further reading.

http://www.slaney.org/pct/pct-toc.html

Thanks for your complete answer. I will check this method and its principles out.

Thanks Marcin for your nice suggestions. I am not sure if thermal scanning works here as the interfaces I am thinking about are not near the surface. It can be like 5 to 10 mm covered with the matrix. Then I cannot be sure if the changes of thermal properties (in this case thermal conductivity) is related to the interface problems or the existing defects in the matrix. 

Hello Mr. Ghiassi,

I do agreed with the suggestion of Mr. Marcin Bielecki. How about the idea with "Thermal wave imaging method" ? The technology deals with the acquisition and analysis of thermal information from non-contact thermal imaging devices. The different techniques involved pulsed, step heating, lock-in, vibrothermography and pulsed phase thermography. Many researchers are using these technologies for identification of hidden defects. Hope these will work for you too.

Dear Shrestha,

Thanks for your complete answer. I think the problem with thermal methods is that the results are sensitive to defects in different depths (meaning is not so straightforward to say the detected defect is at the right place), not to mention the effect of relative humidity, etc. Don't you agree? On the other hand, these methods have a very low resolution and the minimum defect size distinguishable with them is rather large. I would like to use a technique which I can check the possible (small) defects at specific depth. 

Another technique that may be worth looking into is pulsed and transient eddy current analysis. Thomas W. Krause and co-authors have many high impact papers on this analysis in the context of steam generators, gas and oil pipelines and military aircraft wing inspection. I attended a presentation by one of his students on an inspection technique for rivets. 

So far I can think of no method not yet cited.

You definitely must have a closer look to thermal imaging, but in your case with vibroacoustic stimulation (by a sonotrode). I would not be so sure that pulse thermography will not give results, but you will have to choose carefully the source.

Ultrasonic methods by transmission methods may (or may not!) work. It is worth a try with rather low frequencies transducers.

IR thermography might be useful for the evaluation of interface of fibres