*5.4. E*ff*ective Time for Delamination Detection*

In passive IRT, the favorable time in capturing thermal images is a very important factor in order to be applied effectively to real structure inspections. On the rainy day (day 2), almost all delaminations could not be detected owing to the small temperature differences. Therefore, the concrete bridge deck inspection should be carried out on a sunny day.

The detected time of delaminations from the back face on days 1 and 3 is graphically depicted in Figure 19. The blue filled-in square, blue filled-in circle, red unfilled-in square, and red unfilled-in circle represent the first time during daytime, first time during nighttime, last time during daytime, and last time during nighttime when the delaminations can be detected respectively. Under the effect

of the heating cycle during the daytime and cooling cycle during the nighttime, there are two ranges of detectable time when the defects with depths equal to or smaller than 4.0 cm may be detected.

Under the effect of the heating cycle (daytime): The structure inspection should be conducted from 10:00 until 16:00 (6 h) to detect delaminations with WTDRs equal to or higher than 1.9 as shown by the green rectangular in Figure 19. If a delamination with a WTDR of 1.9 is eliminated, the feasible time is from 9:00 to 16:00 (7 h) to identify delamination with the WTDRs equal to or higher than 2.0. However, the inspection of the structure should be carried out from 10:00 to 15:00 (5 h) such that significantly high temperature difference can be obtained, and a clearer observation of delamination can be achieved as shown in Figures 11 and 15.

Under the effect of the cooling cycle (nighttime): If the structure is inspected from 19:30 to 3:30 (8 h), delaminations with the WTDR equal to or higher than 2.8 can be identified. However, in the case of delaminations with WTDRs from 2.5 to 2.8, the suitable collection time is shortened as the period from 19:30 to 21:30 (2 h). In addition, the temperature difference of delaminations achieves a significantly high value from 19:30 to around 2:00 (6.5 h), the structure inspection therefore should be conducted during this time to achieve good results.

In fact, the detected time of the delaminations lasts significantly during both daytime and nighttime. Therefore, the concrete bridge deck inspection can be conducted effectively throughout the day. However, during the daytime, the WTDRs of delaminations that can be detected using the passive IRT technique are smaller in comparison with the nighttime. Thus, daytime is a better choice for concrete bridge inspection than nighttime even if the delaminations with small WTDRs (≤ 2.5) are considered in the inspection. In this case, the limitation of WTDR is approximately 1.9. Conversely, if the inspection focuses only on delamination with WTDRs of 2.8 or higher, the nighttime offers a more efficient selection because of its longer detected time compared to daytime.

**Figure 19.** Feasible time on a sunny day to detect delamination in concrete bridge deck.

## **6. Detectability of Delaminations Using UAV-IRC**

In the case of using the UAV-IRC, the detectability of delamination is also evaluated by using the absolute contrast measurement that is calculated from Equation (3). The temperature difference is the amplitude of the absolute contrast.

Figure 20 shows the thermal image at 11:30 by using UAV-IRC. The temperature difference of delaminations (B-D1 to B-D11) ranges from 5.73 to 0.54 ◦C that is higher than 0.5 ◦C, while it is 0.4 ◦C for the delamination B-D12. The largest WTDR delamination obtains the highest temperature difference (5.73 ◦C) whereas the smallest temperature difference is achieved in the case of the least WTDR delamination (0.4 ◦C). Therefore, all the given delaminations can be seen on the thermal images captured by using UAV-IRC except for delamination B-D12. Hence, it can be concluded that the UAV-IRC might detect delaminations at the depth equal to or lower than 4 cm and with WTDRs not smaller than 1.9 during daytime on the sunny day. However, it should be noted that delamination B-D11 (WTDR = 1.9) is still quite difficult to be recognized as shown in Figure 20.

**Figure 20.** Thermal image of specimen surface using UAV-IRC at 11:30.

Figure 21 shows the absolute contrast of delaminations having different WTDRs on day 5. Particularly, Figure 21a shows the case of defects at the same depth of 2 cm (B-D1, B-D2, B-D3, and B-D4) while Figure 21b depicts defects at depths of 2 (B-D4) and 3 cm (B-D5 and B-D6). The WTDRs of delaminations B-D1, B-D2, B-D3, and B-D4 are 7.9, 7.0, 5.25, and 3.5, respectively, whereas the WTDR is 4.5 and 4.0 corresponding to delaminations B-D5 and B-D6. The same tendency can be observed in comparison with the results obtained from the H-IRC that a larger delamination WTDR achieves higher temperature difference than a smaller one, although it is located at the same depth or further from the surface of the structure. For example, in the case of different WTDRs at the same depth of 2 cm, at 10:30, the temperature differences are 5.22, 3.78, 3.30, and 1.55 ◦C corresponding to delaminations B-D1, B-D2, B-D3, and B-D4. On considering delaminations with different WTDRs and depths, delamination B-D5, B-D6, and B-D4 obtained the temperature difference of 2.70, 2.40, and 1.75 ◦C at 10:30, respectively. Consequently, delamination B-D5 and B-D6 can appear more clearly than B-D4.

The interchange period occurred in the period from 06:30 to 08:30 on day 5 that is very similar to the results from experiments using H-IRC on day 1. It should be noted that the experiments on both days 1 and 5 focus on the back face and the overall weather conditions are sunny during these days. There is only 30 min of the delay time of the interchange period between day 5 and day 1. It might be caused by the difference between the interval in capturing thermal images that are 30 min and 1 h in the case of day 1 and day 5, respectively.

**Figure 21.** Absolute contrast of delaminations with different WTDRs on day 5 from UAV-IRC: (**a**) at same depth of 2 cm; (**b**) at different depths of 2 and 3 cm.

Figure 22 shows the absolute contrast of delamination B-D2 (depth of 2 cm) and B-D5 (depth of 3 cm) based on the results of experiments using UAV-IRC and H-IRC on day 5. In Figure 22a,b, the blue line represents H-IRC while the red line stands for UAV-IRC. At early hours in the morning, the variance of temperature difference between UAV-IRC and H-IRC, known as DIF is much smaller than at other times because the surface temperature in the early morning is more stable than at noon as well as the afternoon, and that time is nearby the interchange period. For example, the DIF is 0.05 ◦C at 5:30 while it is 0.29 ◦C at 12:30 in the case of delamination B-D2. Furthermore, at 11:30, both H-IRC and UAV-IRC obtained the maximum temperature differences. These maximum values are 3.48 ◦C and 2.95 ◦C for B-D5 corresponding to the outcome from H-IRC and UAV-IRC. In addition, it is shown that larger the delamination depth, higher the DIF obtained except for the observation at 9:30 am. For instance, DIFs are 0.29 ◦C and 0.44 ◦C at 12:30 corresponding to delamination B-D2 and B-D5.

Moreover, as shown in Figure 22a,b, both lines show the same tendency, but the red line is located more closely to zero axis compared to the blue line. This can be explained by the effect of the wind from the fans of UAV and the smaller thermal sensitivity of IR camera mounted on the UAV in comparison with H-IRC leading to the decrease in absolute contrast in case of UAV-IRC. However, the gap between the two lines is not significant, which implies that there is a good agreement between the results obtained from experiments using UAV-IRC and H-IRC. Consequently, it can be stated that under the given environmental conditions, using UAV-IRC can give appropriate results. The effective application of UAV-IRC in the defect detection of concrete bridge decks is confirmed.

**Figure 22.** Comparison between the absolute contrast captured by H-IRC and UAV-IRC on a sunny day (day 5): (**a**) delamination B-D2; (**b**) delamination B-D5.
