A 2D-FFT signal processing technique to enhance the contrast of air-coupled ultrasonic C-scans

摘要

In this work, a Fourier-based algorithm to enhance the contrast of air-coupled ultrasonic C-scans is presented. The performances of the developed algorithm are checked on C-scans recorded in order to monitor the state of conservation of ancient paintings. Experiments on simulated and real ancient paintings have confirmed the ability of the algorithm in recognizing small corrupted regions, replacing them with synthetic values and finally filtering out the high frequency noise. The processed images show an enhanced contrast and an higher signal to noise ratio and appear pretty easy to be interpreted.