Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Jul 13;23(14):6362.
doi: 10.3390/s23146362.

Generative Deep Learning-Based Thermographic Inspection of Artwork

Yi Liu  1 , Fumin Wang  1 , Zhili Jiang  1 , Stefano Sfarra  2 , Kaixin Liu  3 , Yuan Yao  4
Affiliations

Generative Deep Learning-Based Thermographic Inspection of Artwork

Yi Liu et al. Sensors (Basel). .

Abstract

Infrared thermography is a widely utilized nondestructive testing technique in the field of artwork inspection. However, raw thermograms often suffer from problems, such as limited quantity and high background noise, due to limitations inherent in the acquisition equipment and experimental environment. To overcome these challenges, there is a growing interest in developing thermographic data enhancement methods. In this study, a defect inspection method for artwork based on principal component analysis is proposed, incorporating two distinct deep learning approaches for thermographic data enhancement: spectral normalized generative adversarial network (SNGAN) and convolutional autoencoder (CAE). The SNGAN strategy focuses on augmenting the thermal images, while the CAE strategy emphasizes enhancing their quality. Subsequently, principal component thermography (PCT) is employed to analyze the processed data and improve the detectability of defects. Comparing the results to using PCT alone, the integration of the SNGAN strategy led to a 1.08% enhancement in the signal-to-noise ratio, while the utilization of the CAE strategy resulted in an 8.73% improvement.

Keywords: artwork; convolutional autoencoder; generative adversarial network; infrared thermography; panel painting.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
IRT detection device.
Figure 2
Figure 2
Framework of the GPCT model.
Figure 3
Figure 3
Framework of the CPCT model.
Figure 4
Figure 4
Madonna sample and defect distribution: (a) Madonna sample; (b) defect distribution diagram.
Figure 5
Figure 5
Original thermal images, where the grid division for F-score calculation is illustrated in the first sub-figure.
Figure 6
Figure 6
Results of image enhancement based on deep learning strategies: (a) generated images; (b) enhanced images.
Figure 7
Figure 7
Comparison of different methods: (a) PCT; (b) SPCT; (c) GPCT; (d) CPCT.
Figure 8
Figure 8
Comparison of the SNR values of the five rounds of test results of the proposed model with the PCT results: (a) GPCT; (b) CPCT.
See this image and copyright information in PMC

References

    1. Garrido I., Erazo-Aux J., Lagüela S., Sfarra S., Ibarra-Castanedo C., Pivarčiová E., Gargiulo G., Maldague X., Arias P. Introduction of deep learning in thermographic monitoring of cultural heritage and improvement by automatic thermogram pre-processing algorithms. Sensors. 2021;21:750. doi: 10.3390/s21030750. - DOI - PMC - PubMed
    1. Giorgi R., Baglioni M., Berti D., Baglioni P. New methodologies for the conservation of cultural heritage: Micellar solutions, microemulsions, and hydroxide nanoparticles. Acc. Chem. Res. 2010;43:695–704. doi: 10.1021/ar900193h. - DOI - PubMed
    1. Kudinov I.I., Golovkov A.N., Shishkin P.A., Skorobogatko D.S., Andreev A.I., Generalov A.S. Evaluating the efficiency of using ultraviolet radiation sources in carrying out fluorescent penetrant testing. Russ. J. Nondestruct. Test. 2022;58:57–69. doi: 10.1134/S1061830922010041. - DOI
    1. Cheng X., Ma G., Wu Z., Zu H., Hu X. Automatic defect depth estimation for ultrasonic testing in carbon fiber reinforced composites using deep learning. NDT E Int. 2023;135:102804. doi: 10.1016/j.ndteint.2023.102804. - DOI
    1. Fernandes H., Summa J., Daudre J., Rabe U., Fell J., Sfarra S., Gargiulo G., Herrmann H.G. Characterization of ancient marquetry using different non-destructive testing techniques. Appl. Sci. 2021;11:7979. doi: 10.3390/app11177979. - DOI

LinkOut - more resources