An in situ high temperature microwave microscope was built for detecting surface and sub-subsurface structures and defects. This system was heated with a self-designed quartz lamp radiation module, which is capable of heating to 800°C. A line scanning of a metal grating showed a super resolution of 0.5 mm (λ/600) at 1 GHz. In situ scanning detections of surface hole defects on an aluminium plate and a glass fiber reinforced plastic (GFRP) plate were conducted at different high temperatures. A post processing algorithm was proposed to remove the background noises induced by high temperatures and the 3.0 mm-spaced hole defects were clearly resolved. Besides, hexagonal honeycomb lattices were in situ detected and clearly resolved under a 1.0 mm-thick face panel at 20°C and 50°C, respectively. The core wall positions and bonding width were accurately detected and evaluated. In summary, this in situ microwave microscope is feasible and effective in sub-surface detection and super resolution imaging at different high temperatures.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Near-field scanning microwave microscope (NSMM)  has become a powerful tool for super resolution imaging and noninvasive characterization of various materials such as dielectrics [2–4], semiconductors [5,6] and metals [7,8]. Synge  firstly described the concept to surpass the Abbe barrier using an opaque screen with a small sub-wavelength diameter hole (10 nm in diameter), held about 10 nm above the surface of a smooth flat sample. Ash and Nichols’s seminal paper  demonstrated a NSMM using a quasi-optical hemispherical resonator and obtained a resolution of λ/60 at 10 GHz. Since then, different kinds of NSMMs have been developed based on a resonator [3,11] or a transmission line [12–14]. For microscopes based on cavities or lumped element resonators, limited by the specific resonance structure, they can only be operated at a single frequency or a severe narrow frequency band. While for microscopes based on coaxial transmission line resonators, a broad band can be achieved since the resonance mode structure is maintained over a wide frequency range. Local dielectrics properties have been imaged in a broad band from 1.3 to 17.4 GHz using the microwave microscope of this type . Furthermore, microwave microscopes based on an interferometry technique have been proposed offering both broadband capabilities and high measurement sensitivities [16,17]. This technique has been applied for broadband dielectric characterization in liquid media [18,19] and epitaxial graphene , and even vital mitochondria in respiration buffer has been imaged using a interferometric-based reflectometer .
Besides broadband and high sensitivity measurement, high spatial resolution is another pursuit in the realm of microwave microscope. The spatial resolution of a near-field scanning microwave microscope is governed by the probe size since the evanescent field is confined in the vicinity of the probe apex. Nanometer resolution and even atom scale resolution is available through using sharp probe tips [22,23]. However, there is a contradiction between the spatial resolution and the scan range. When a microwave microscope with micron resolution is used to scan an area of several square millimeters or even several square centimeters, it will be extremely time-consuming. Therefore, there should be a compromise between the resolution and the detection range in nondestructive testing.
On the other hand, the need for nondestructive testing (NDT) systems working in high temperature environments has been rising. In petroleum chemical industry, high temperature pressure vessels and pipes  commonly suffer from corrosion , wall-thinning , etc., which may finally induce leakage or explosion. Conventional NDT methods are usually limited to be applied until the components have been cooled to room temperature, and it takes a lot of time and seriously damages the economic benefits. While on-line or in-service high temperature NDT system will save time and energy, lower the cost and improve efficiency. Besides, high temperature NDT systems are desperately in need in the realm of hypersonic flight. With the rapid development of hypersonic vehicles, the high temperature environment induced by aerodynamic heating is becoming more and more severe. Therefore, studies of thermal protection materials and structures of hypersonic vehicles have become a key issue [27–30]. In high temperature conditions, variations of shapes, properties, and defect initiation and evolution will significantly degrade the structural and functional integrity of thermal protection materials. To better understand the thermal behaviors of these special materials in a hot environment, high temperature simulation experiments are necessary. Therefore, it is urgent and important to build experimental apparatuses for in situ high temperature nondestructive detections.
Herein, we self-built an in situ high temperature microwave NDT system capable to work from room temperature to 800°C. A thermally protected coaxial connector probe was designed and manufactured using beryllium copper. The broadband reflection spectra were measured at different high temperatures. In situ high temperature microwave nondestructive detections were performed on surface pit holes of an aluminium plate and a glass fiber reinforced plastic (GFRP) plate at different temperatures. An image processing algorithm was proposed to solve the non-uniform background illumination problem induced by thermal-induced deformation and torsion of specimens. A hexagonal honeycomb lattices under a 1.0 mm-thick face panel was detected with super resolution at 20°C and 50°C, respectively.
2. Experimental setup
Figure 1 depicts the self-built in situ high temperature microwave NDT system, which consists of a radiation heating module, a home-made near-field microwave probe and a room temperature microwave NDT setup . The room temperature microwave NDT setup is capable to realize automatic scanning detections with a computer simultaneously controlling the PNA to transmit and receive microwaves and the x-y-z scanning stage to move. In , a coaxial connector probe was used to detect liquid ingress defect in the honeycomb core-layer, and a super resolution was obtained by overcoming the edge diffraction effects. Herein, the home-made probe is protected with heat insulation and water cooling, which is designed to enable it suitable for in situ detections at high temperatures. The details of each module are as follows.
The radiation heating module was designed using an array of quartz lamps capable of heating to 800°C. Eight quartz lamps were arranged in parallel with their separation distance optimized to provide a relatively uniform thermal field. They were all in parallel connection with the temperature controller through a switch, which can individually control the on/off state of each quartz lamp. The heating cavity was designed open. Thick heat-insulated mullite slabs with heating windows of different sizes could be chosen to cover the cavity according to the size of the specimen under test. This assured an effective heating on the target area. The bottom surface of the cavity was specially designed and installed with a polished high temperature alloy plate, which can enhance the thermal reflection. The interlinings of the cavity walls were filled with mullite slabs and heat insulation cottons to keep the near outside of the cavity a room temperature environment.
A near-field microwave microscope probe was designed based on a 2.92 mm connector, whose operating frequency is from DC to 40 GHz. Figure 2(a) is the design drawing of the probe. As a compromise of the sub-millimeter resolution and the scan range of several square millimeters, the diameter of the cylindrical central conductor is designed in the range of sub-millimeter, as 0.6 mm. In comprehensive consideration of sparing enough length for water cooling and avoiding the flexibility of long probe, the length of the outer conductor is designed as 15.0 mm. The fringe electric field in the vicinity of the probe in free space was simulated at 1 GHz, as shown in Fig. 2(b). It clearly demonstrates that the evanescent field is confined at the end of the probe. Besides, the electric field distribution on a metal (i.e. a perfect conductor) surface beneath the probe tip was simulated at 1 GHz at a standoff distance of 0.5 mm, as shown in Fig. 2(c). The full width at half maximum is about the size of the probe tip, which implies a resolution capability of about 0.6 mm. The electric field maximum of the illuminated spot is a function of the standoff distance, normalized |S11| decreases sharply with the increasing standoff distance in the range of 0~2 mm, as shown in Fig. 2(d). This can be qualitatively analyzed by the circuit model of the tip-sample interaction, as depicted by the inset in Fig. 2(d). The capacitance of the tip-sample separation layer is inversely proportional to the standoff distance, and this capacitance dominates the near-field tip-sample interaction. Therefore, a standoff distance value can be chosen in 0~2 mm in consideration of the sensitivity and energy intensity.
The reflection coefficient () detected by the near-field microwave microscope is related to the impedance () distribution of the material under test:
The probe was manufactured with beryllium copper, which can withstand a high temperature up to 930°C. The insertion of the connection part was polyether imide (PEI) and polytetrafluoroethylene (PTFE), which can work under 165°C. Therefore, to protect the probe from high temperature, a water cooling system was designed with a thin water-cooling tube wound round the central conductor of the probe. Besides, the coaxial cable and the connection section were wrapped tightly with aerogel felt materials as thermal shields. The probe was fixed on the scanning table through a rigid beam, while the specimen was fixed on the windowed mullite slab. Raster scanning detections through moving the probe over the specimen would be conducted to image amplitude and phase of the reflection coefficient (S11). An electronic calibration module was used to calibrate the system errors of the PNA and the coaxial cable.
Four K-type thermocouples were arranged as illustrated in Fig. 1 to measure the temperatures of the cavity near the bottom surface of the specimen, the top surface of the specimen, and the probe tip and the probe after water cooling. The latter two thermocouples attached to the probe were removed in scanning detection to avoid interferences to the probe.
3. Results and discussion
3.1 Heating test
An aluminium plate with thickness of 5.0 mm was used as the specimen in heating test. A mullite slab with a 10 cm × 10 cm window was chosen to cover the cavity. The distance between the probe tip and the specimen surface was adjusted to 0.5 mm. Four K-type thermocouples were mechanically fixed in the cavity near the bottom surface of the specimen, on the top surface of the specimen, on the probe tip and on the central conductor part above the water cooling section, respectively. An objective temperature of the cavity was set as 500°C with a temperature-control precision of 0.1°C. In case of serious overshoots, only the two central quartz lamps right below the heating window were turned on.
As shown in Fig. 3, four heating curves were recorded with the multichannel thermodetector. It takes about 3 minutes for the cavity temperature to rise from room temperature to 200°C. The heating rate decreases with the increase of the temperature. It takes about 25 minutes to reach up to 500°C. The top surface temperature of the aluminium plate rises approximatively linearly to about 210°C. The temperatures of the probe tip and the probe part after water cooling rise slowly along with the cavity temperature. The effect of water cooling system becomes more obvious at higher temperature, and finally a temperature decrease of about 30°C is obtained. It implies that water cooling is an effective way to decrease the probe temperature and enable the near-field probe to work at higher temperatures.
3.2 Near-field probe test
A metal grating with line and gap width of 0.5 mm was machined as shown in the inset in Fig. 4(a). Both amplitude and phase of S11 were measured at 1 GHz (λ = 300 mm) and the grating was clearly resolved. It verified that this near-field microwave microscope was capable of obtaining a super resolution of λ/600.
The high temperature performance of this home-made microscope probe was investigated through measuring the reflection coefficients of an aluminium plate (10 cm × 20 cm) at different high temperatures. Figure 4(b) shows the reflection spectra in frequency band of 1~40 GHz (781 points) at 20°C, 100°C, 200°C, 300°C, 400°C, 500°C, respectively. The standoff distance was set as 0.5 mm. The input power was 0 dBm, and the intermediate frequency bandwidth (IFBW) was set as 1 kHz. Considering the signal-to-noise ratio in detection, the amplitude of S11 over 0.5 can be adopted, which corresponds to frequency bands of 1~5 GHz, 7~13 GHz and 16.5~22 GHz, and frequency points at about 30 GHz and 40 GHz. Therefore, this probe is available to be operated in multi-bands. The overall trend of these reflection spectra were in consistent, while the variations become more obvious at higher temperatures. It implies that high temperatures affect received microwaves in detection. The reflection spectra at different high temperatures were averaged, and the average curve along with high temperature noise envelope was plotted in Fig. 4(c). The high temperature noise envelope becomes wider at relative high frequencies, which indicates that high temperatures affects more serious at relative high frequencies.
To evaluate the effect of the high temperature and the temperature gradient of the probe on the detected reflection coefficient, both the heating curves in Fig. 3 and reflection spectra in Fig. 4(b) were used. As shown in Fig. 3, both the probe tip temperature and the temperature difference between the probe tip and the probe after water cooling are increasing along with the cavity temperature. When the cavity temperature reaches up to 500°C, the probe tip temperature and the temperature difference of the probe reaches up to their maximums of about 100°C and 30°C, respectively. The absolute relative errors of |S11| between 20°C and 500°C were calculated and plotted in Fig. 4(d), as shown on the left axis. Simultaneously, the reflection spectrum at 20°C is shown on the right axis. The scale values on the left axis was set one-tenth of those on the right axis. The horizontal dashed line corresponds to |S11| equal to 0.5. It clearly shows that in the multi-bands of about 1~5 GHz, 7~13 GHz and 16.5~22 GHz, and at about 30 GHz and 40 GHz, where |S11| is more than 0.5, the absolute values of |S11| between 20°C and 500°C is much less than one-tenth of |S11| at 20°C. It implies that in this temperature range, the probe tip temperature and temperature gradient of the probe affect little on the variation of |S11| in these multi-bands.
3.3 In situ detection at high temperatures
Case1: Blind holes on an aluminium plate
Two 3.0 mm-spaced holes with the same diameter of 1.5 mm were machined on an aluminium plate. This metal specimen was attached on the windowed mullite plate covering the heating window with the two holes in the center. Firstly, it was scanned at room temperature and the amplitude and phase of S11 at frequency of 1 GHz were obtained and mapped, as shown in Fig. 5(a) and 5(b). The two surface hole defects are resolved more clearly in the phase mapping than in the amplitude one. A 6 dB drop method was applied in the phase mapping to evaluate the hole size and separation. As shown in Fig. 5(c), the dashed circles with diameter of 1.5mm (λ/200) and separation of 3.0 mm (λ/100) are in accordance with the processed phase image. As to the amplitude imaging, it shows more susceptible to the diffraction effect from hole sides, displaying two large and blurry spots. Besides, background noises affect more seriously in amplitude mapping. The background noises attribute to the large scanning area with invisible surface tilt and unevenness. For a microscope of nanometer-scale resolution, the surface of the detection area is approximately flat, while for a macro-scale microwave microscope, the surface tilt or unevenness becomes prominent over the detected large area. Therefore, for a large detecting area, the background noises should be post-processed, and amplitude and phase information should be compared to each other in the evaluation of defects.
In situ detections of this metal specimen were conducted at different high temperatures. Figure 6(a) shows the scanning images at 100°C (cavity temperature near the bottom surface of the specimen) at 1 GHz, 10 GHz, 20 GHz, 30 GHz, 40 GHz, respectively. The top row is amplitude and the bottom row is phase. Similarly, amplitude and phase images at 200°C, 300°C, 400°C, and 500°C were obtained, as shown in Fig. 6(b). Overall, the phase images are much clearer than amplitude ones, while at 40 GHz the quality of amplitude images is better than that of phase ones. It implies the advantage of a multi-band probe over single frequency type that complementary amplitude or phase information may be obtained at additional frequency. It also shows that the quality of scanning images degrades with the increasing of temperature. This attributes to the increasing variation of standoff distance over the scanning area caused by surface tilt and torsion at high temperatures. Therefore, in order to resolve the hole defects more clearly, the blurred images should be post processed.
Herein, we proposed a correction algorithm for non-uniform background illumination. The basic idea of this algorithm is to evaluate and reconstruct the distribution of the background noises, and then subtract it from the raw image, at last enhance the image through normalization. The core of this algorithm is based on two morphological transformations of white Top-Hat (WTH) and black Top-Hat (BTH) [32,33], which are defined as
Take the amplitude image at room temperature as an example, as shown in Fig. 5(a), corresponding background image, background removed image and the enhanced image were shown in Fig. 7(a)-7(c), respectively. The two line distributions of the normalized amplitude through the hole centers, as indicated by the dashed lines in Fig. 7 (c), were extracted and plotted in Fig. 7 (d). Diameters of the two hole defects are quantitatively evaluated as 1.6 mm and 1.4 mm, respectively. It demonstrates that the diameter of the hole defect was accurately evaluated as 1.5 ± 0.1 mm, which shows a super resolution of λ/200. Besides, the two hole defects are clearly resolved with the background noises removed in the corrected and enhanced image compared to the blurred raw one. Therefore, it verifies that this algorithm is effective in removing background noises and accurate for quantitative evaluation of defect sizes.
Case 2: Blind holes on a GFRP laminate
Composite laminates are usually exposed to high temperature environments. We also conduct in situ noncontact detections of surface holes on a 1.7 mm-thick GFRP laminate. The diameter and center-to-center separation were 1.5 mm and 3.0 mm, respectively. The standoff distance was set as 0.5 mm. Amplitude and phase scanning images were obtained at 30 GHz at 20°C, 50°C, 100°C and 150°C. As shown in Fig. 8 (a), the raw images of amplitude and phase also suffer from non-uniform background noises, which degrade the image quality with blurs. The post-processing algorithm described in Case 1 was applied to remove these background noises, and the processed images were accordingly shown in Fig. 8 (b). After post processing, the surface hole defects were clearly resolved. It confirms that non-uniform background illumination, which was induced by the variation of standoff distance due to specimen tilt and torsion, is a common problem for both metal and composite specimen at room temperature and different high temperatures. This post processing algorithm was verified effective in removing non-uniform background noises at high temperatures.
Case 3: Hexagonal honeycomb lattice
As described in the simulation part of this probe, the effective detection range in air of this probe is about 0~2 mm. It implies a larger detection depth in dielectric specimen than nano-scale resolution microscope. Therefore, it is attractive to verify the depth detection capability of this probe in experiment. A honeycomb sandwiched plate was manufactured with the top panel in thickness of 1.0 mm, and the side length and the wall thickness of the hexagonal honeycomb core cell were 3.0 mm and 0.1 mm, respectively. The standoff distance was set as 0.5 mm.
Figure 9 (a) shows the amplitude and phase mapping of the inner hexagonal honeycomb lattice at room temperature and at 50°C at 1 GHz. Though the imaging quality was degraded at high temperature with additional noises, the hexagonal lattice is clearly characterized in all the scanning images. It indicates that this microwave microscope probe is capable of detecting sub-surface structures under depth in millimeter-scale. Figure 9 (b) shows the phase distribution on the dashed line as indicated in Fig. 9 (a). Positions of the core walls are clearly detected, as the small peaks show at 2.17 mm and 7.38 mm, respectively. Therefore, the spacing between opposite core walls is 5.21 mm, which is consistent with the actual spacing of 5.2 mm calculated according to the side length of 3.0 mm. Figure 9 (c) shows the phase mapping at room temperature processed through Canny arithmetic operators, and boundaries of the adhesive bonding are clearly resolved. The width of the bonding section is about 1.2 mm (λ/250). It demonstrates that this probe is effective in accurately detecting the core wall positions and evaluation of the bonding width with a super resolution. Therefore, this near-filed microscope is feasible and effective in sub-surface detection and super resolution imaging at high temperatures.
A high temperature microwave NDT system was built with a designed quartz lamp radiation module as the heater. A home-made microwave microscope probe was manufactured and assembled in the system for in situ high temperature detections. It was verified available in multi-bands for subsurface detection and super resolution imaging at high temperatures. A metal grating with line and gap width of 0.5 mm was clearly resolved at 1 GHz, therefore, a λ/600 super resolution was obtained. In situ detections of surface holes on an aluminium plate and a GFRP laminate were performed at different high temperatures, a post processing algorithm was proposed and verified effective in solving the non-uniform background illumination problem. Honeycomb cores under a 1.0 mm-thick face panel were clearly resolved through in situ high temperature detection, which verified the sub-surface detection capability of this system. Besides, the core wall positions and bonding thickness were accurately detected and evaluated. Therefore, this in situ high temperature microwave NDT system is feasible and effective in sub-surface detection and super resolution imaging at high temperatures.
National Natural Science Foundation of China (NSFC) (11522214, 11521202); the Key Subject “Computational Solid Mechanics” of China Academy of Engineering Physics.
References and links
1. A. Imtiaz, T. M. Wallis, and P. Kabos, “Near-field scanning microwave microscopy: An emerging research tool for nanoscale metrology,” IEEE Microw. Mag. 15(1), 52–64 (2014). [CrossRef]
2. A. Karbassi, D. Ruf, A. D. Bettermann, C. A. Paulson, D. W. van der Weide, H. Tanbakuchi, and R. Stancliff, “Quantitative scanning near-field microwave microscopy for thin film dielectric constant measurement,” Rev. Sci. Instrum. 79(9), 094706 (2008). [CrossRef] [PubMed]
3. C. Gao, B. Hu, I. Takeuchi, K. S. Chang, X. D. Xiang, and G. Wang, “Quantitative scanning evanescent microwave microscopy and its applications in characterization of functional materials libraries,” Meas. Sci. Technol. 16(1), 248–260 (2005). [CrossRef]
4. A. P. Gregory, J. F. Blackburn, K. Lees, R. N. Clarke, T. E. Hodgetts, S. M. Hanham, and N. Klein, “Measurement of the permittivity and loss of high-loss materials using a Near-Field Scanning Microwave Microscope,” Ultramicroscopy 161, 137–145 (2016). [CrossRef] [PubMed]
5. A. N. Reznik, I. A. Shereshevsky, and N. K. Vdovicheva, “The near-field microwave technique for deep profiling of free carrier concentration in semiconductors,” J. Appl. Phys. 109(9), 145–148 (2011). [CrossRef]
6. S. Berweger, J. C. Weber, J. John, J. M. Velazquez, A. Pieterick, N. A. Sanford, A. V. Davydov, B. Brunschwig, N. S. Lewis, T. M. Wallis, and P. Kabos, “Microwave Near-Field Imaging of Two-Dimensional Semiconductors,” Nano Lett. 15(2), 1122–1127 (2015). [CrossRef] [PubMed]
7. J. Rossignol, C. Plassard, E. Bourillot, O. Calonne, M. Foucault, and E. Lesniewska, “Non-destructive technique to detect local buried defects in metal sample by scanning microwave microscopy,” Sens. Actuators A Phys. 186, 219–222 (2012). [CrossRef]
8. L. You, J. J. Ahn, Y. S. Obeng, and J. J. Kopanski, “Subsurface imaging of metal lines embedded in a dielectric with a scanning microwave microscope,” J. Phys. D Appl. Phys. 49(4), 45502 (2015). [CrossRef]
9. E. H. Synge, “XXXVIII. A suggested method for extending microscopic resolution into the ultra-microscopic region,” Lond. Edinb. Dublin Philos. Mag. J. Sci. 6(35), 356–362 (1928). [CrossRef]
11. J. Kim, M. S. Kim, K. Lee, J. Lee, D. Cha, and B. Friedman, “Development of a near-field scanning microwave microscope using a tunable resonance cavity for high resolution,” Meas. Sci. Technol. 14(1), 7–12 (2003). [CrossRef]
12. C. P. Vlahacos, R. C. Black, S. M. Anlage, A. Amar, and F. C. Wellstood, “Near‐field scanning microwave microscope with 100 μm resolution,” Appl. Phys. Lett. 69(21), 3272–3274 (1996). [CrossRef]
13. N. Qaddoumi, M. A. Khousa, and W. Saleh, “Near-field microwave imaging utilizing tapered rectangular waveguides,” in IEEE Instrum. Meas. Technol. Conf. (2004), pp. 174–177. [CrossRef]
14. S. M. Anlage, V. V. Talanov, and A. R. Schwartz, “Principles of near-field microwave microscopy,” in Scanning Probe Microscopy (Springer, 2007), pp. 215–253.
16. K. Haddadi and T. Lasri, “Broadband Microwave Interferometry for Nondestructive Evaluation,” in 13th International Symposium on Nondestructive Characterization of Materials (NDCM-XIII) (2013), pp. 10–16.
17. H. Bakli, K. Haddadi, and T. Lasri, “Interferometric technique for scanning near-field microwave microscopy applications,” IEEE Trans. Instrum. Meas. 63(5), 1281–1286 (2014). [CrossRef]
18. K. Haddadi, S. Gu, and T. Lasri, “Sensing of liquid droplets with a scanning near-field microwave microscope ☆,” Sens. Actuators A Phys. 230, 170–174 (2015). [CrossRef]
19. S. Gu, T. Lin, and T. Lasri, “Broadband dielectric characterization of aqueous saline solutions by an interferometer-based microwave microscope,” Appl. Phys. Lett. 108(24), 242903 (2016). [CrossRef]
20. S. Gu, X. Zhou, T. Lin, H. Happy, and T. Lasri, “Broadband non-contact characterization of epitaxial graphene by near-field microwave microscopy,” Nanotechnology 28(33), 335702 (2017). [CrossRef] [PubMed]
21. J. Li, Z. Nemati, K. Haddadi, D. C. Wallace, and P. J. Burke, “Scanning Microwave Microscopy of Vital Mitochondria in Respiration Buffer,” arXiv Prepr. arXiv1802.05939 (2018).
22. A. Imtiaz, S. M. Anlage, J. D. Barry, and J. Melngailis, “Nanometer-scale material contrast imaging with a near-field microwave microscope,” Appl. Phys. Lett. 90(14), 143106 (2007). [CrossRef]
23. J. Lee, C. J. Long, H. Yang, X. Xiang, and I. Takeuchi, “Atomic resolution imaging at 2. 5 GHz using near-field microwave microscopy,” Appl. Phys. Lett. 97(18), 5–7 (2010). [CrossRef]
24. P. F. Mastro, Pressure Vessels and Pipes (John Wiley & Sons, Inc., 2016).
25. S. Haladuick and M. R. Dann, “Risk-Based Maintenance Planning for Deteriorating Pressure Vessels With Multiple Defects,” J. Press. Vessel Technol. 139(4), 41602 (2017). [CrossRef]
26. X. Du, Y. Liu, and J. Zhang, “High Temperature Limit Analysis of Pressure Vessels and Piping with Local Wall-Thinning,” (2018). [CrossRef]
27. Y. Yang, J. Yang, and D. Fang, “Research progress on thermal protection materials and structures of hypersonic vehicles,” Appl. Math. Mech. 29(1), 51–60 (2008). [CrossRef]
28. D. Glass, “Ceramic Matrix Composite (CMC) Thermal Protection Systems (TPS) and Hot Structures for Hypersonic Vehicles,” in 15th AIAA International Space Planes and Hypersonic Systems and Technologies Conference (2008), p. 2682. [CrossRef]
29. E. A. Thornton, “Thermal structures-four decades of progress,” J. Aircr. 29(3), 485–498 (1992). [CrossRef]
30. C. Zhou, Z. Wang, and T. Hou, “Heat Transfer Analysis of Thermal Protection Structures for Hypersonic Vehicles,” IOP Conf. Ser. Mater. Sci. Eng. 269(1), 12020 (2017). [CrossRef]
31. P. Wang, Y. Pei, and L. Zhou, “Near-field microwave identification and quantitative evaluation of liquid ingress in honeycomb sandwich structures,” NDT Int. 83, 32–37 (2016). [CrossRef]
32. G. Wang, Y. Wang, H. Li, X. Chen, H. Lu, Y. Ma, C. Peng, Y. Wang, and L. Tang, “Morphological background detection and illumination normalization of text image with poor lighting,” PLoS One 9(11), e110991 (2014). [CrossRef] [PubMed]
33. R. C. Gonzalez and R. E. Woods, “Digital image processing,” Prentice Hall Int. 28(4), 484–486 (2005).