Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
20 Laboratory and Field Evaluations of Infrared Thermography Systems The chapter was prepared by Kenneth Maser of Infrasense, Inc. Collection of Infrared Data Two infrared (IR) cameras were made available for this test- ing: (a) an Infrared Cameras, Inc. (ICI), 7320 camera and (b) a FLIR A40M camera with a wide-angle lens. The ICI 7320 had a resolution of 320 à 240 pixels and a frame rate of 30 Hz. The FLIR A40M also had the same pixel resolution but a maximum frame rate of 60 Hz. The wide-angle lens was used in conjunction with the FLIR camera so that a full lane width could be imaged from a mounting platform about 13 ft above the pavement surface. ICI offers a similar wide- angle lens, but this lens was not available at the time of testing. The primary difference between the two cameras is size. The FLIR camera weighs about 3 lb and measures 3 in. à 3 in. à 6 in. The ICI camera is about the size of a pocket digital camera. Laboratory Evaluation For each test slab, the IR cameras were operated from a ladder to provide sufficient height for a complete IR image view of the entire test slab. A series of still, IR images were recorded at specified times during the heating/cooling cycle of the test slab. The slabs were heated by using an array of eight high- intensity heat lamps, and surface and bottom slab tempera- ture were continuously monitored during the heating process. A photograph of this heating setup is shown in Figure 3.1. During the heating process, the area around the heated slab was enclosed to contain the heat, as shown in Figure 3.2. After the slabs were heated, IR measurements were made with both the FLIR and ICI cameras from a ladder, with each camera about 11 to 12 ft above the floor. The height was nec- essary to obtain images that included the full extent of the test slab. The initial testing was carried out with the heat- ing lamps positioned at approximately 2 ft above the slab. The initial results showed a heating pattern that emulated the bulb array, as shown in Figure 3.3. The presence of this pat- tern prevents the detection of thermal anomalies associated with subsurface defects. As a result, the bulbs were raised to the maximum height possible in the test setup, which was about 3 ft above the surface of the slabs. The resulting heating pattern was more uniform. Figure 3.4 shows thermal anoma- lies in Slab B. Field Evaluation at Test Track For this field testing, Infrasense provided a mounting and recording system that consisted of the following: ⢠A camera-mounting frame for elevating each IR camera above a survey vehicle approximately 13 ft above the pave- ment surface, with motorized, remotely controlled pan/tilt head for precise camera positioning; ⢠A video camera and digital video recorder for collecting standard visual images of the pavement surface in parallel with the IR images; ⢠A vehicle wheel-mounted distance measuring instrument for encoding distance on the video images and for trigger- ing the collection of data from the IR cameras; and ⢠Laptop-based data-acquisition software. The physical setup of the IR data-collection system is shown in Figure 3.5. Testing using the equipment shown in Figure 3.5 was carried out continuously by using a vehicle speed of approximately 3 mph. During each test, IR images were collected at 1-ft intervals and sequentially stored on the laptop hard drive. Simultaneously, a visual video image was recorded to a digital video recorder. The distance traveled was encoded with a counter and superimposed on the video image using a video overlay device. A sample pair of infrared and cor- responding video images is shown in Figure 3.6. Note that because the video image was collected at the normal NTSC rate of 30 frames/s, the encoded distance on the video image C h a p t e r 3
21 Figure 3.1. Heating of test slabs with heat lamps. Figure 3.2. Test slab enclosure during heating process. Figure 3.3. IR image of Slab A showing bulb pattern. (a) (b) Figure 3.4. Image of Slab B showing thermal anomalies (a) immediately after heating and (b) 45 minutes after removal of heat.
22 FLIR Camera on Pan-Tilt Head ICI Camera Figure 3.5. IR thermography setup for track testing. (a) (b) Figure 3.6. Corresponding IR and video images: (a) video image with encoded distance and (b) IR image. provided distance registration. The corresponding IR image is Image 143, corresponding to 143 ft of travel. The initial set of IR system tests was carried out Novem- ber 8, 2009. Two series of tests were carried outâone from 1 to 2 p.m. and the second from 3 to 4 p.m. The tempera- ture conditions on the track pavement were relatively the same during each series of tests. The weather conditions were sunny with air temperatures of around 60°F. The second series of tests was carried out March 7 and 8, 2010. One of the objectives of this second series was to explore a larger range of temperatures. During this second series, tests were carried out at the times and ambient temperatures listed in Table 3.1. There was no significant cloud cover during this second series of tests. Note that the range of pavement test temperatures ranged from 35°F to 103°F. Infrared Data analysis Laboratory Evaluation The laboratory IR data were visually observed and quali- tatively correlated with the known defects in the test slab. Figure 3.7 shows thermal images of Slabs A and B taken immediately after heating. Thermal anomalies are observed on both slab images. For Slab A, the near anomaly is located in the intact section, and thus is apparently caused by factors other than subsurface delamination (e.g., uneven heating). The anomalies in Slab B could be related to the debonding and stripping at 4-in. depth, but the results of Slab A suggest that these anomalies Table 3.1. Thermal Properties of Components in Each Model Test Time Ambient Temperature (°F) Pavement Temperature (°F) 2:00 p.m. 63 NA 4:00 p.m. 60 NA 8:30 p.m. 42 NA 6:00 a.m. 34 35 9:00 a.m. 60 70 10:00 a.m. 60 80 11:30 a.m. 60 95 12:45 p.m. 62 103 Note: NA = not available.
23 could also be due to other factors. Therefore, the laboratory IR tests were inconclusive. Field Evaluation at Test Track The IR images from the test track test section were spliced together to produce a single composite strip IR image for the entire test section. One such image was produced for each test. The images were evaluated visually, in conjunction with the visual video data, to identify IR anomalies that could be associ- ated with subsurface debonding, delamination, and stripping. Figure 3.8 shows an example of the analysis of the IR data. The figure shows a portion of the composite IR image rep- resenting Sections 4, 5, 6, and 7 at the test track. The image shows some anomalies in Section 6 in the area where 0.75 in. of reclaimed asphalt pavement was placed 2 in. down from the surface to represent stripping. However, examination of the surface video shows that these anomalies correspond with surface features and tire marks rather than with subsurface features. In general, the IR image anomalies did not clearly correlate with the known subsurface defects in the test section. (a) (b) Figure 3.7. Test slab images immediately after heating for (a) Slab A and (b) Slab B. yellow line rumble strip stripping areas Sect 4 Sect 5 Sect 6 Sect 7 surfacing transition Figure 3.8. IR strip image of Sections 4 through 7.
