To comprehensively assess the effectiveness of infrared thermal imaging technology in detecting delamination defects in indoor wall plaster layers, this study designed specimens with varying defect sizes, thicknesses, and plasters. These specimens were divided into two main categories: one using cement mortar and the other using gypsum mortar as the plastering material. To ensure that the number and area of defects met the research requirements, each specimen had dimensions set at a length of 1,240 mm, a width of 1,220 mm, and a thickness of 220 mm, and they were constructed using aerated concrete blocks. To closely simulate real-world application scenarios, the delamination defects were intentionally placed between the plastering layer and the wall. Detasils regarding the size, thickness, and quantity of these defects are provided in Figure 1. Polyurethane foam, with a thermal conductivity coefficient close to that of air (0.022 W/m·°C), was chosen as the simulated material for the defects.

To simulate wall construction in actual engineering as closely as possible, the specimens were constructed using staggered masonry, following the relevant construction standards outlined in “Technical Specification for Acceptance of Construction Quality of Masonry Structures” GB 50203–2011. Initially, a utility knife was used to cut the polyurethane foam boards to the designed dimensions. Subsequently, they were sanded with sandpaper to achieve the precise dimensions of polyurethane foam required to simulate the form of plaster delamination defects. According to the specimen’s design requirements, the polyurethane foam was fixed to the front of the masonry specimen before plastering, as shown in Figure 2. These steps were taken to ensure that the construction of the specimen closely resembled wall construction in actual engineering.

The plasters used in this study include cement mortar and gypsum mortar, both of which are widely used in actual engineering. The proportion of cement mortar consists of cement, sand, and water in a ratio of 1:3:0.6, with ordinary Portland cement used as the cement type. On the other hand, the gypsum mortar proportion consists of gypsum, sand, and water in a ratio of 1:2.5:0.5. When plastering the specimens, the front side of each specimen was designed to have plastered surfaces with delamination defects, while the reverse side had standard plastered surfaces, as shown in Figure 3. Figure 3A displays the cement mortar plastered surface, while Figure 3B shows the gypsum mortar plastered surface. To ensure a plaster layer thickness of 10mm, 10 mm steel nails were inserted into the masonry structure before plastering, and the plaster thickness was determined with reference to these steel nails during the plastering process. The selection of a 10 mm thickness for plaster layers in the work is justified by its ability to adequately cover wall imperfections, enhance structural durability, and improve insulation and soundproofing. This thickness is ideal for providing protection and increasing comfort. The quality of plastering complies with the requirements of “Code for Acceptance of Construction Quality of Building Decoration Engineering” GB50210-2018.2. In the preparation of gypsum mortar, use ordinary fine sand and construction-grade gypsum powder, mixing them in a ratio of gypsum to sand to water of 1:2.5:0.5. This should be done quickly and evenly with an electric mixer to prevent premature hardening and lumps. The mixture must be applied immediately and evenly. The process for cement mortar is similar, involving a mix of cement and sand added to water and blended for immediate application on a prepared surface.

To simulate a typical indoor environment of a building, this experiment was conducted inside a civilian building located in Yinying Village, Zhenping County, Nanyang City, Henan Province, China. The specific arrangement of the experiment is shown in Figure 4A. Data collection was performed using the TI395 infrared thermal imager (Figure 4B), whose performance specifications are presented in Table 1. Considering the impact of the indoor environment on imaging, the infrared thermal imager was placed at a distance of 4.5 m directly in front of the specimen (Figure 4A). This placement ensures the integrity of the specimen’s imaging in both the vertical and horizontal directions.

This study consisted of two rounds of experiments, each lasting for 48 h and commencing upon completion of plastering work on the specimens to collect data. The infrared thermal imager used was configured to automatically capture thermal images every 45 s, resulting in a cumulative total of 3,840 thermal images for each round of the experiment. Simultaneously, the temperature and humidity data logger RC-4HA was set to record environmental temperature and humidity data every 15 min, starting after the plastering was completed and continuing for 48 h. The equipment setup used in the experiment and the trends in environmental temperature and humidity can be observed in Figures 4C, D.

To calculate the surface temperature of the specimens using radiometry, initial parameters needed to be set in the infrared camera before data collection. These parameters included ambient temperature, ambient humidity, emissivity of the plaster surface, and the distance from the camera to the test object. Based on these parameters, the camera automatically calculated the atmospheric transmittance. According to ASTM standards (Tran et al., 2017), it is not recommended to conduct such tests when wind speeds exceed 24 km/h (American Society for Testing and Materials, 2007). Given that the indoor wind speed was close to zero, the conditions for this experiment were suitable. Furthermore, as this experiment was not influenced by direct solar radiation, heat transfer primarily occurred through convection with the surrounding air, making environmental temperature a critical factor. To focus on analyzing delamination defects in the plaster layer, wooden boards were placed in the experiment to minimize the interference of edge effects.

Despite the inherent challenges in selecting reference and defect areas, the absolute contrast method, also known as thermal or absolute thermal contrast, continues to be the predominant approach for the analysis of thermal image sequences (Kretzmann et al., 2016; Tran et al., 2018b). This method is particularly valued for its straightforward and effective means of quantitatively evaluating the defect detection capabilities of infrared thermography (IRT) technology, as well as for analyzing the optimal time window for defect detection. Despite the ASTM D4788-03 standard stating that defects are detectable only when the absolute contrast between the defect and healthy areas exceeds 0.5°C, practical evidence indicates that lower temperature differences are also sufficient for defect identification (Huh et al., 2016; Kretzmann et al., 2016; Hiasa et al., 2017; Rocha and Póvoas, 2017; Mac et al., 2019). Hence, considering practical outcomes and the need for higher precision in certain applications, the work has chosen ±0.2°C as our threshold for detecting defects.

Figure 5 delineates the patterns of absolute contrast variations within layered defect regions across various specimens. Specifically, Figure 5A illustrates the behavior in cement mortar specimens, while Figure 5B focuses on gypsum mortar specimens (Defects are represented by capital letters in Figure 5B). In the case of cement mortar, Figure 5C demonstrates that layered defects were successfully identified using IRT, a success attributed to the absolute contrast in these defect areas exceeding the 0.2°C threshold within a predetermined timeframe. It is important to note that the absolute contrast within the cement mortar’s stratified defect regions displayed a complex pattern characterized by an initial decrease, followed by an increase, and then a series of decreases and increases, culminating in a final decrease. Additionally, variations were observed in the absolute contrast across different defect areas. Conversely, Figure 5D presents the detection of various stratified defect regions within gypsum mortar specimens, showcasing an absolute contrast trend that diverges from that observed in cement mortar specimens. This trend is characterized by an initial decrease followed by an increase, with notable differences in the absolute contrast among different defect areas being observed as well.

To delve deeper into the relationship between the absolute contrast of layered defect areas and the size and thickness of defects, and thereby analyze the threshold conditions for detecting plaster layer defects using infrared thermography, it was observed that the absolute contrast in the defect areas of both cement and gypsum mortar specimens showed irregular patterns within the initial 17 and 14 h, respectively. Hence, the latter, more regular half of the experiment was chosen for analysis, selecting the time frames of 34 h for cement mortar specimens and 26 h for gypsum mortar specimens, to curve-fit the relationship between absolute contrast and defect size and thickness, as shown in Figure 6. Figures 6A, B reveal that for cement mortar specimens, the relationship between absolute contrast and thickness for defects of sizes 20 cm, 15 cm, and 10 cm, as well as the relationship between absolute contrast and the square of the size for defects of thicknesses 5 mm, 3 mm, and 1mm, approximate a linear relationship, as indicated by the R-Square fitting evaluation indices. For gypsum mortar materials, Figures 6C, D demonstrate that the absolute contrast for defects of thicknesses 5, 3, and 1 mm, and the size to the fourth power, as well as the absolute contrast for defects of sizes 20, 15, and 10 cm and the square of their thickness, also exhibit a linear relationship, with R-Square values very close to 1. These analyses indicate that, under the conditions of this experiment, for cement mortar materials, the increase in absolute contrast gradually grows with the increase in defect size under the same thickness conditions; similarly, under the same size conditions, the absolute contrast uniformly increases with the increase in defect thickness. The same patterns apply to gypsum mortar materials: with the increase in size under the same thickness conditions, the increase in absolute contrast gradually grows; likewise, under the same size conditions, the increase in defect thickness leads to a gradual increase in absolute contrast. These findings provide a significant quantitative basis for detecting plaster layer defects using infrared thermography technology.

Considering the relatively fixed thickness of indoor plastering, there exists a certain linear relationship between the depth of defects and their thickness. This allows for the depth factor to be considered secondary in the detection of indoor plastering delamination defects, focusing instead on thickness and size as the primary detection metrics. This study introduces the concept of Width-to-Thickness Ratios (WTTRs)—the ratio between the size of the delamination and its thickness—as a measure for analyzing the detectability of delamination defects. This concept has been widely applied in the implementation of infrared thermal imaging technology for concrete bridge decks, as well as in exploring the impact of solar radiation on imaging subsurface features in concrete, to analyze the detectability of layered defects.

Figure 7 presents the detectability analysis of delamination defects. Based on the approximate linear relationship between the thickness and size of the plastering (cement and gypsum) delamination defects and their absolute contrast discussed in the previous section, a certain detection range has been confirmed. In this experimental setup, all nine defects were effectively detected within this range. Given the findings that larger delamination defect thickness and size result in higher absolute contrast, thus favoring defect detection, it can be reasonably inferred that within 48 h post-plastering, as long as the delamination defects exceed a thickness of 1 mm, size of 100 mm, and have a width-to-thickness ratio between 20 and 200, these defects can be effectively detected.

Determining an effective detection time window is crucial for the application of passive infrared thermography technology in on-site structural inspections. By analyzing the time-domain absolute contrast curves of layered defects caused by hydration reactions and environmental temperature changes, this study identifies the corresponding detection time windows, as illustrated in Figure 8.

Figure 8 reveals two detection time windows for layered defects in cement mortar plasters. The first window, lasting from 8 to 9 h, allows for the observation of defects immediately after the start of data collection. This could be attributed to the rise in environmental temperature causing the plaster surface to heat up and enabling early detection of layered defects. The second detection window extends from 14 to 29 h, with 3 mm and 15 × 15 mm² defects first observed at 20.3 h. Defects measuring 1 mm and 10 × 10 mm² are no longer observable after 36.5 h, yet defects such as 5 mm and 20 × 20 mm², 3 mm and 20 × 20 mm², 1mm and 20 × 20 mm², and 1 mm and 15 × 15 mm² remain detectable at 48 h. Notably, the 3 mm and 10 × 10 mm² defect deviates from this pattern, lasting only 2.4 h, but the absolute contrast approaches the threshold between 22.9 and 36 h.

For gypsum mortar plasters, as shown in Figure 9, two detection time windows are also observed. The first window is shorter, lasting between 2 and 3.9 h, with 1 mm and 15 × 15 mm² defects initially observed at 4.2 h and 5 mm and 10 × 10 mm² defects becoming undetectable after 9.5 h. The second window, lasting from 25 to 38 h, allows for the observation of defects such as 5 mm and 15 × 15 mm², 5 mm and 20 × 20 mm², 3 mm and 20 × 20 mm², and 5 mm and 15 × 15 mm², all of which remain detectable at 48 h.

Given the significant impact of plastering material on the detection time windows, this study sets the detection window for cement mortar plasters from 22 to 42 h, with the optimal detection time point at 34 h. For gypsum mortar plasters, the detection window is set from 6 to 8 h, with 7 h as the optimal detection time point. These periods not only enable the detection of all layered defects but also encompass the time points where the absolute contrast peaks.

The impact of ambient temperature on infrared thermographic detection results is significant, especially when detecting delamination defects in cement and gypsum mortar specimens. In Figure 9, This work observed the changes in absolute contrast of the delamination defect areas measuring 5mm, 200 × 200mm², along with the variations in ambient temperature, through the data collection process for specimens of two different materials.

For cement mortar specimens, Figure 10A clearly reveals the consistency between the trend of absolute contrast changes in the delamination defect areas and the trend of ambient temperature changes. Notably, the peak values of absolute contrast for delamination defects within two detection time windows almost perfectly align with the peak areas of ambient temperature, underscoring the critical role of ambient temperature in the detection of delamination defects in cement mortar specimens during the curing stage.

However, the situation is slightly different for gypsum mortar specimens, as shown in Figure 10B. Initially, there is a lack of consistency between the changes in absolute contrast of the delamination defect areas and the changes in ambient temperature, and the peak values of absolute contrast for delamination defects in the initial detection window do not align with the peak values of ambient temperature. This indicates that ambient temperature is not the dominant factor affecting the detection of delamination defects in gypsum mortar specimens in the early stages. During the 0–6 h of temperature decline, the hydration heat of the gypsum mortar partially offset by the environmental cooling is detrimental to the detection of delamination defects. However, in the later stages, as the release of hydration heat decreases, ambient temperature becomes a facilitating factor, allowing the detection of some delamination defects, thereby indicating that ambient temperature plays a leading role in the detection of delamination defects in gypsum mortar specimens during the curing stage in the later phases.