In the CIMDS Visual Intelligent Sensor, Measurement, and Control Lab, the goal of our aging-aircraft inspection research is to use robots' help to find cracks and corrosion on aircraft fuselages. We are currently adding computer image enhancement and automated image understanding to the raw stereo video imagery. Meanwhile we believe that with proper sensor/data fusion, the coin-tap test can be a great help to achieve our goal.
The coin tap test (or "screwdriver handle test") is a venerable means for manually verifying the integrity of objects and structures, particularly sheet-like and layered materials that are subject to cracking and delamination. Healthy examples typically reverberate cleanly ("live"), whereas damaged examples yield a sound that is dull ("dead").
Our survey of the literature in this field, the commercial products now on the market, the instruments being used in key applications (aircraft skins, boat hulls), and our own experiment results tend to support our working hypothesis that microphones and accelerometers have their separate valid roles as instrumentation suitable for automating defect detection via the "coin tap test".
Our research improves our understanding of the fundamental principles underlying the individual measurement techniques, and leads to developing an understanding of their correlation, particularly as it relates to possibilities of making the "whole greater than the sum of the parts" via sensor fusion.
For more detailed information please see my paper "Correlation of Accelerometer and Microphone Data in Coin Tap Test"; in postscript or pdf format.