*4.2. Experimental Setup*

The passive IRT test was conducted at Chonnam National University, Republic of Korea (Figures 7 and 8a). Two IR detectors were employed in this study, i.e., an H-IRC (FLIR, Wilsonville, OR, USA) and a UAV-IRC (FLIR, Wilsonville, OR, USA) (Figure 8b,c and Figure 9). The H-IRC is a long-wavelength handheld IR camera (FLIR SC660) in the 7.5–13.5 μm bandwidth [49]. This camera has a focal plane array of 640 × 480 pixels, thermal sensitivity ≤ 0.03 ◦C (at + 30 ◦C), spatial resolution (IFOV) of 0.65 mrad, field of view (FOV) of (24◦ × 18◦)/0.3 m, and an accuracy of ±1 ◦C. It should be noted that long-wavelength cameras in the 7.5–13.5 μm range work well when atmospheric attenuation is involved because the atmosphere tends to act as a high-pass filter above 7.5 μm. The H-IRC is located at a height of 195 cm and the distance from the closest edge of the specimen to the camera lens is 360 cm that allows capturing the temperature of the whole specimen surface in only one thermal image (Figure 8b). Overall, the surface temperature was captured at intervals of 30 min.

**Figure 7.** Location of the test.

Another long-wavelength IR camera (FLIR Zenmuse XT2 - FLIR, Wilsonville, OR, USA) with the same bandwidth as that of the H-IRC is mounted on the UAV-IRC DJI M200 version [50]. The thermal sensitivity of the UAV-IRC is equal to or less than 0.05 ◦C, and the FOV is 57.12◦ × 42.44◦. The technical data of the UAV is shown in Table 2 [50].

**Figure 8.** Arrangement of the experiment: (**a**) location of specimen; (**b**) with H-IRC; (**c**) with UAV-IRC.

**Table 2.** Main technical data of the DJI M200 version UAV.


All experiments except the case of using UAV-IRC were conducted on both faces of the specimen to consider depths from 2 to 7 cm. The passive IRT test was conducted for five days. The collection time was from 6:00 to 6:00 (24 h), 8:30 to 24:00 (15 h and 30 min), 9:00 am to 21:00 pm (12 h), 8:00 to 8:00 (24 h), and 5:30 to 13:30 (8 h) on the 1st, 2nd, 3rd, 4th, and 5th days respectively.

The scope of each experiment cycle is indicated in Table 3. The thermal images were captured only by the H-IRC during the first four days whereas both H-IRC and UAV-IRC were used on day 5. The capability of the UAV-IRC in detecting delaminations in the concrete deck is analyzed by comparing it with respect to the obtained data from the experiments using H-IRC.

The relative humidity and ambient temperature are important parameters that the camera must be fed with before capturing thermal images. Then, the surface temperature is automatically calculated by the IR camera (FLIR, Wilsonville, OR, USA) using Equation (2) in Section 3. Therefore, the ambient temperature, relative humidity as well as wind speed were measured by the Kestrel 3000 (Nielsen-Kellerman, Chester, PA, USA) before taking thermal images. The equipment used in this study is shown in Figure 9.



**Figure 9.** Equipment used in the experiment: (**a**) H-IRC; (**b**) Kestrel 3000; (**c**) UAV-IRC.
