Wavelet filtering of NDT signals

Separating useful signals from noise – effective quality assurance methods

At the turn of the millennium, wavelets were nothing more than a new method for mathematicians to effectively process signals at the time-frequency level. The groundbreaking mathematics were soon used for efficient compression algorithms in digital photography. However, the use of wavelet filter technology to solve NDT-related problems turned out to be extremely promising.

Typical applications include increasing the signal-to-noise ratio and the separation of the different signal components of non-stationary signals. We have been working with this technology for a full 15 years now and were the first to apply real-time methods based on relevant electronic hardware (DSP and FPGA) in NDT operations for pipe production. This includes the processing of “slower” signals from magnetic flux leakage testing and eddy current testing as well as “fast” signals from ultrasonic testing. For all applications, dedicated systems are available in in-house development which can, as a rule, be looped into the analog signal path. As a result, it is also possible to equip existing inspection systems with this efficient filter technology.