Accuracy on the Time-of-Flight Estimation for Ultrasonic Waves Applied to Non-Destructive Evaluation of Standing Trees: A Comparative Experimental Study

Abstract

Time-of-flight measurement is a critical step to per-2 form ultrasonic non-destructive testing of standing 3 trees, with direct influence on the precision of de-4 fect detection. Aiming to increase the accuracy on 5 the estimation, the characteristics of the ultrasonic 6 measurement chain should be adapted to the con-7 straints of wood testing in living condition. This 8 study focused on the excitation signal parameters, 9 such as shape, temporal duration, and frequency re-10 sponse, and then the selection of a suitable time-11 of-flight determination technique. A standing plane 12 tree was tested, placing ultrasonic receivers at four 13 different positions, with five different excitation sig-14 nals and three time-of-flight detection methods. The 15 proposed ultrasonic chain of measurement resulted in 16 high signal-to-noise ratios in received signals for all 17 configurations. A time-frequency analysis was used 18 to determine the power distribution in the frequency 19 domain, showing that only chirp signal could concen-20 trate the power around the resonant frequency of the 21 sensor. Threshold and Akaike information criterion 22 method performed similar for impulsive signals with 23 decreasing uncertainty as sensor position approached 24 to the radial direction. Those two methods failed to 25 accurate determine time-of-flight for Gaussian pulse 26 and chirp signals. Cross-correlation was only suitable 27 for the chirp signal, presenting the lower uncertainty 28 values among all configurations.

Ultrasound computed tomography on standing trees: accounting for wood anisotropy permits a more accurate detection of defects

“…Additionally, two electronic amplifiers were used, one located after the signal generator (Single Channel High Voltage Linear Amplifier A800, FLC Electronics) to increase the energy sent to the US transmitter, and other at the US receiver output (Physical Acoustics Corporation AE2A/AE5A), both with an amplification of 40 dB. The excitation signal was a chirp signal (central frequency of 60 kHz, bandwidth of 48 kHz), leading to a concentrated power spectrum around the sensor central frequency that allows a time-of-flight (TOF) measurement using a cross-correlation (Espinosa et al 2018b). The full set of TOF measurements (i.e., projections) was obtained by placing the transmitter at the 16 different points along the wood disk circumference.…”

“…For the FBP image, the mean velocity was 1551 m/s with σ = 315 m/s. Even when both images presented relatively flat surfaces, the FBP image presented more artifacts near to the borders, where the precision of the TOF measurements is lower (Espinosa et al 2018b). Two horizontal profiles traversing the center of the image are shown in Fig.…”

Ultrasonic imaging of standing trees: factors influencing the decay detection

“…The number of sensors, for example, is directly related to the spatial resolution of the image; a larger number of sensors allows a smaller defective area detection [11,12]. The excitation signal used for testing is directly related to the precision of the TOF estimation, affecting the image reconstruction method [13,14]. This image reconstruction method must be adapted to the wood anisotropy condition, in order to avoid the bias introduced by the isotropic hypothesis considering straight-line trajectories [15,16].…”

Effect of Location, Clone, and Measurement Season on the Propagation Velocity of Poplar Trees Using the Akaike Information Criterion for Arrival Time Determination

“…Traditionally, the ToF method uses the threshold technique to determine the arrival time of the signal [3,17,23]. Alternatively, other methods like noise-reduction-based methods [24], higher-order statistics-based methods [25], floating threshold, Akaike information criterion [26], etc. have been proposed in the literature, for arrival time determination of transient signals.…”