Color crosstalk correction for synchronous measurement of full-field temperature and deformation

https://doi.org/10.1016/j.optlaseng.2021.106878Get rights and content

Highlights

  • The origin of color crosstalk of high temperature multi-channel synchronous measurement is analyzed in detail.

  • Developing a novel method to realize the synchronous measurement of temperature and deformation while addressing the challenge of color crosstalk.

  • Establishing a set of correction equations to obtain the crosstalk-eliminated light intensity of each channel.

  • Synchronous measurement of temperature and deformation is accomplished with the thermal heating experiment up to 1100 °C.

Abstract

Synchronous measurement of temperature and deformation at elevated temperatures based on the noncontact optical method can be achieved through the allocation of channels R, G, and B for a color camera. Channels R and G are used for the radiation-based temperature measurement, while channel B is adopted for the deformation measurement. However, the existence of color crosstalk, which is mainly caused by the color camera's spectral response and the use of the narrow-band filtering technique, has a significant influence on the accuracy of temperature and deformation calculations. In this work, the color crosstalk effect in high-temperature optical imaging is analyzed, and a crosstalk correction algorithm based on grayscale calibration is performed on the different color channels (channels R, G, and B) of the color image. By solving the calibration equations, three corrected color channels are generated for a high-precision temperature and deformation calculation. A flame heating experiment of a C/SiC material up to 1100 °C is carried out to validate the applicability accuracy of the proposed method. The results show that the proposed method can effectively eliminate the color crosstalk between different channels and provide a non-contact, full-field, and synchronous measurement of temperature and deformation with higher accuracy than the existing methods.

Introduction

High-temperature structural materials, including superalloys, ceramics, intermetallics, and composites, have played an important role in aerospace engineering and gas turbines [1], [2], [3], [4], [5]. It is of great significance to evaluate the key parameters, such as the strain and temperature, at elevated temperatures, which provides an important reference for material selection, component design, and structure optimization [6,7]. The issue of high-accuracy, full-field, and multiparameter measurements in extreme environments has received considerable critical attention. Limited by the temperature resistance of materials, the requirements of material mechanics evaluation at high temperatures (e.g., over 800 °C) usually cannot be satisfied with conventional contact measurement methods. Recently, the noncontact measurement method based on optical imaging has been widely used owing to its advantages of full-field, long-term measurement in a wide temperature range [8], [9], [10], [11].

For strain measurement, the digital image correlation (DIC) method has been widely used to measure the full-field strain over the surface of the specimen [12], [13], [14], [15], [16], [17]. Combined with the optical bandpass filtering technique, the DIC measurement has been extended to a high temperature over 2000 °C [18]. Meanwhile, temperature measurements have also drawn increasing concern. On the one hand, a large number of material properties, such as the elastic modulus and strength of materials, are temperature-dependent. On the other hand, the directly obtained strain is the coupling of mechanical strain and thermal strain, which need to be further decoupled by the temperature field. Thus, considerable literature has been published around the theme of synchronous measurement of temperature and deformation [19], [20], [21].

For a high-temperature object, the radiated light and the reflected light contain the temperature and deformation information of the surface, respectively. The color camera, which is most commonly used as the optical imaging system, includes channels red (R), green (G), and blue (B), which can be separated for further signal processing using a single image. Su et al. [19] proposed an experimental setup with a channel allocation strategy using an optical technique to synchronously measure the full-field temperature and deformation at high temperature using only one camera, which significantly reduced the complexity of the experiment. The deformation measurement was based on the high-temperature DIC method, and the temperature field was calculated using an improved two-color method. Their experiment on a C/SiC composite tested at a temperature up to 1100 °C showed great potential for simultaneous temperature and deformation measurements. However, optical channel crosstalk may inevitably occur for a commonly used color camera since the 3-color camera is too expensive to be applied in engineering tests [22,23]. Therefore, the measurement accuracy based on the light intensity of channels R, G, and B will be significantly affected.

In principle, the derivation of the improved two-color method is based on Planck's blackbody radiation theory, which requires that the light intensity be purely obtained from the object's radiation to ensure the accuracy of the temperature measurement. However, with the influence of channel crosstalk, the responses of channels R and G to the blue light band are not zero, as are the responses of channels B and G to the red light band. Meanwhile, external blue light illumination combined with a blue bandpass filter is usually adopted to suppress excessive thermal radiation for better imaging quality. The spectrum of the blue illumination is approximately a Gaussian distribution with a central peak wavelength in the blue band and bandwidths ranging from a few nanometers to tens of nanometers, which will further exacerbate the effects of blue light on channel crosstalk, especially for a relatively low temperature [20]. This influence can be reduced by eliminating initial light intensity [24], which, however, may fail with the requirements of adjusting exposure time and aperture to adapt to varied thermal radiation. In addition, compared with the CCD camera, the filtered pixel array of the CMOS camera similarly responds to a broader spectrum. It is more commonly used in engineering applications due to cost, spatial resolution, and the requirement of dynamic measurement. Therefore, the key issue for the high-precision and synchronous measurement of temperature and deformation is the extraction of single-channel sub-images without color crosstalk from the collected color images.

The main object of this work is to develop a novel method to realize the synchronous measurement of temperature and deformation while addressing the challenge of color crosstalk. External narrow-band red, green, and blue lights were adopted to illuminate the sample surface, while the exact interactive influence of the three channels was evaluated by functions using the least-squares fit method. Thus, a set of correction equations was established and solved to obtain the crosstalk-eliminated light intensity of each channel. Synchronous measurement of temperature and deformation was accomplished with the thermal heating experiment of the C/SiC sample, which indicated that the proposed method can effectively reduce the influence of crosstalk.

Section snippets

Principles for synchronous measurement of temperature and deformation

Here, we briefly describe the principles for synchronous measurement of the full-field temperature and deformation at high temperature, which was first proposed by Su et al. [19]. Note that the deformation measurement is realized by reflected light, and the radiation light carries the temperature information of an object's surface. According to Planck's blackbody radiation theory, the radiation is the strongest within the red (R) wavelengths, followed by the green (G) wavelengths, and weakest

Determination of calibration functions

It is reasonable to assume that color crosstalk is caused by the spectral response characteristics of the camera itself because of the irradiation of a narrowband light source. The elimination of color crosstalk includes two steps: calibration and correction. In the calibration step, the imaging characteristics of the camera are obtained under irradiation with a monochromatic or narrowband light source. It can be described by a set of calibration functions indicating intensity values of

Experimental setup

To verify the effectiveness of the proposed method, we carried out a high-temperature experiment of heating the C/SiC sample using an oxypropane flame. It should be noted that color crosstalk depends on camera imaging characteristics rather than temperature so that the calibration function can be obtained at room temperature. The experimental setup is illustrated in Fig. 7, which mainly includes a CMOS color camera (SP-5000C-USB, JAI), a lens with a focal length of 50 mm, an infrared pyrometer

Conclusion

We propose a color crosstalk correction method based on the imaging characteristics to reduce the contamination for channels R, G, and B in one camera and further improve the high-temperature optical technique for synchronous measurement of the full-field temperature and deformation. The method was validated by experiments on a C/SiC composite subjected to flame heating up to 1100 °C. The experimental results show that the proposed method can effectively reduce the intensity caused by crosstalk

CRediT authorship contribution statement

Mengkun Yue: Conceptualization, Visualization, Methodology, Formal analysis, Writing – original draft, Writing – review & editing, Resources. Jinyang Wang: Methodology, Formal analysis, Writing – original draft, Writing – review & editing. Jinsong Zhang: Methodology, Formal analysis, Writing – original draft, Writing – review & editing. Yao Zhang: Methodology, Resources. Yunlong Tang: Conceptualization, Visualization. Xue Feng: Conceptualization, Visualization, Supervision.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

We gratefully acknowledge financial support from the National Natural Science Foundation of China (Grant Nos. U20A6001, 11625207, 11921002, and 12102401).

References (24)

  • M.D. Novak et al.

    High-temperature materials testing with full-field strain measurement: experimental design and practice

    Rev Sci Instrum

    (2011)
  • T. Cai et al.

    Simultaneous measurement of two-dimensional temperature and strain fields based on thermographic phosphor and digital image correlation

    Meas Sci Technol

    (2021)
  • Cited by (0)

    View full text