Abstract

Image spectroscopy (IS) is an important tool for the noninvasive analysis of works of art. It generates a wide sequence of multispectral images from which a reflectance spectrum for each imaged point can be recovered. In addition, digital processing techniques can be employed to divide the images into areas of similar spectral behavior. An IS system designed and developed in our laboratory is described. The methodology used to process the acquired data integrates spectral analysis with statistical image processing: in particular, the potential of principal-component analysis applied in this area is investigated. A selection of the results obtained from a sixteenth-century oil-painted panel by Luca Signorelli is also reported.

© 1998 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. P. J. Ready, P. A. Wintz, “Information extraction, SNR improvement, and data compression in multispectral imagery,” IEEE Trans. Commun. COM-21, 1123–1131 (1973).
    [CrossRef]
  2. P. Geladi, H. Grahn, Multivariate Image Analysis (Wiley, New York, 1996).
  3. D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).
  4. P. R. Norton, “Infrared image sensors,” Opt. Eng. 30, 1649–1663 (1991).
    [CrossRef]
  5. A. Burmester, F. Bayerer, “Towards improved infrared reflectograms,” Stud. Conserv. 38, 145–154 (1993).
    [CrossRef]
  6. G. Thomson, The Museum Environment (Butterworth-Henemann, Stoneham, Mass., 1994).
  7. C. M. Bastuscheck, “Correction of video camera response using digital techniques,” Opt. Eng. 26, 1257–1262 (1987).
    [CrossRef]
  8. K. V. Mardia, J. T. Kent, J. M. Bibby, Multivariate Analysis (Birnbaum and Lukacs, London, 1979).
  9. H. Martens, T. Næs, Multivariate Calibration (Wiley, New York, 1989).
  10. M. Bacci, S. Baronti, A. Casini, F. Lotti, M. Picollo, O. Casazza, “Nondestructive spectroscopic investigations on paintings using optical fibers,” Mater. Issues Art Archaeol. III 267, 265–282 (1992).
  11. R. Linari, M. Picollo, B. Radicati, “Schede di caratterizzazione di pigmenti usati in pittura,” Tech. Rep. TR/POE/92.7 [Istituto di Ricerca sulle Onde Elettromagnetiche (IROE), Firenze, Italy, July1992].

1996 (1)

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

1993 (1)

A. Burmester, F. Bayerer, “Towards improved infrared reflectograms,” Stud. Conserv. 38, 145–154 (1993).
[CrossRef]

1992 (1)

M. Bacci, S. Baronti, A. Casini, F. Lotti, M. Picollo, O. Casazza, “Nondestructive spectroscopic investigations on paintings using optical fibers,” Mater. Issues Art Archaeol. III 267, 265–282 (1992).

1991 (1)

P. R. Norton, “Infrared image sensors,” Opt. Eng. 30, 1649–1663 (1991).
[CrossRef]

1987 (1)

C. M. Bastuscheck, “Correction of video camera response using digital techniques,” Opt. Eng. 26, 1257–1262 (1987).
[CrossRef]

1973 (1)

P. J. Ready, P. A. Wintz, “Information extraction, SNR improvement, and data compression in multispectral imagery,” IEEE Trans. Commun. COM-21, 1123–1131 (1973).
[CrossRef]

Bacci, M.

M. Bacci, S. Baronti, A. Casini, F. Lotti, M. Picollo, O. Casazza, “Nondestructive spectroscopic investigations on paintings using optical fibers,” Mater. Issues Art Archaeol. III 267, 265–282 (1992).

Baronti, S.

M. Bacci, S. Baronti, A. Casini, F. Lotti, M. Picollo, O. Casazza, “Nondestructive spectroscopic investigations on paintings using optical fibers,” Mater. Issues Art Archaeol. III 267, 265–282 (1992).

Bastuscheck, C. M.

C. M. Bastuscheck, “Correction of video camera response using digital techniques,” Opt. Eng. 26, 1257–1262 (1987).
[CrossRef]

Bayerer, F.

A. Burmester, F. Bayerer, “Towards improved infrared reflectograms,” Stud. Conserv. 38, 145–154 (1993).
[CrossRef]

Bibby, J. M.

K. V. Mardia, J. T. Kent, J. M. Bibby, Multivariate Analysis (Birnbaum and Lukacs, London, 1979).

Burmester, A.

A. Burmester, F. Bayerer, “Towards improved infrared reflectograms,” Stud. Conserv. 38, 145–154 (1993).
[CrossRef]

Buydens, L.

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

Cammann, K.

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

Casazza, O.

M. Bacci, S. Baronti, A. Casini, F. Lotti, M. Picollo, O. Casazza, “Nondestructive spectroscopic investigations on paintings using optical fibers,” Mater. Issues Art Archaeol. III 267, 265–282 (1992).

Casini, A.

M. Bacci, S. Baronti, A. Casini, F. Lotti, M. Picollo, O. Casazza, “Nondestructive spectroscopic investigations on paintings using optical fibers,” Mater. Issues Art Archaeol. III 267, 265–282 (1992).

Feldhoff, R.

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

Geladi, P.

P. Geladi, H. Grahn, Multivariate Image Analysis (Wiley, New York, 1996).

Grahn, H.

P. Geladi, H. Grahn, Multivariate Image Analysis (Wiley, New York, 1996).

Huth-Fehre, T.

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

Kantimm, T.

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

Kent, J. T.

K. V. Mardia, J. T. Kent, J. M. Bibby, Multivariate Analysis (Birnbaum and Lukacs, London, 1979).

Linari, R.

R. Linari, M. Picollo, B. Radicati, “Schede di caratterizzazione di pigmenti usati in pittura,” Tech. Rep. TR/POE/92.7 [Istituto di Ricerca sulle Onde Elettromagnetiche (IROE), Firenze, Italy, July1992].

Lotti, F.

M. Bacci, S. Baronti, A. Casini, F. Lotti, M. Picollo, O. Casazza, “Nondestructive spectroscopic investigations on paintings using optical fibers,” Mater. Issues Art Archaeol. III 267, 265–282 (1992).

Mardia, K. V.

K. V. Mardia, J. T. Kent, J. M. Bibby, Multivariate Analysis (Birnbaum and Lukacs, London, 1979).

Martens, H.

H. Martens, T. Næs, Multivariate Calibration (Wiley, New York, 1989).

Melssen, W.

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

Næs, T.

H. Martens, T. Næs, Multivariate Calibration (Wiley, New York, 1989).

Norton, P. R.

P. R. Norton, “Infrared image sensors,” Opt. Eng. 30, 1649–1663 (1991).
[CrossRef]

Picollo, M.

M. Bacci, S. Baronti, A. Casini, F. Lotti, M. Picollo, O. Casazza, “Nondestructive spectroscopic investigations on paintings using optical fibers,” Mater. Issues Art Archaeol. III 267, 265–282 (1992).

R. Linari, M. Picollo, B. Radicati, “Schede di caratterizzazione di pigmenti usati in pittura,” Tech. Rep. TR/POE/92.7 [Istituto di Ricerca sulle Onde Elettromagnetiche (IROE), Firenze, Italy, July1992].

Radicati, B.

R. Linari, M. Picollo, B. Radicati, “Schede di caratterizzazione di pigmenti usati in pittura,” Tech. Rep. TR/POE/92.7 [Istituto di Ricerca sulle Onde Elettromagnetiche (IROE), Firenze, Italy, July1992].

Ready, P. J.

P. J. Ready, P. A. Wintz, “Information extraction, SNR improvement, and data compression in multispectral imagery,” IEEE Trans. Commun. COM-21, 1123–1131 (1973).
[CrossRef]

Thomson, G.

G. Thomson, The Museum Environment (Butterworth-Henemann, Stoneham, Mass., 1994).

van den Broek, W.

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

Wienke, D.

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

Winter, F.

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

Wintz, P. A.

P. J. Ready, P. A. Wintz, “Information extraction, SNR improvement, and data compression in multispectral imagery,” IEEE Trans. Commun. COM-21, 1123–1131 (1973).
[CrossRef]

Anal. Chem. (1)

D. Wienke, W. van den Broek, W. Melssen, L. Buydens, R. Feldhoff, T. Huth-Fehre, T. Kantimm, F. Winter, K. Cammann, “Near-infrared imaging spectroscopy (NIRIS) and image rank analysis for remote identification of plastics in mixed waste,” Anal. Chem. 354, 823–828 (1996).

IEEE Trans. Commun. (1)

P. J. Ready, P. A. Wintz, “Information extraction, SNR improvement, and data compression in multispectral imagery,” IEEE Trans. Commun. COM-21, 1123–1131 (1973).
[CrossRef]

Mater. Issues Art Archaeol. III (1)

M. Bacci, S. Baronti, A. Casini, F. Lotti, M. Picollo, O. Casazza, “Nondestructive spectroscopic investigations on paintings using optical fibers,” Mater. Issues Art Archaeol. III 267, 265–282 (1992).

Opt. Eng. (2)

C. M. Bastuscheck, “Correction of video camera response using digital techniques,” Opt. Eng. 26, 1257–1262 (1987).
[CrossRef]

P. R. Norton, “Infrared image sensors,” Opt. Eng. 30, 1649–1663 (1991).
[CrossRef]

Stud. Conserv. (1)

A. Burmester, F. Bayerer, “Towards improved infrared reflectograms,” Stud. Conserv. 38, 145–154 (1993).
[CrossRef]

Other (5)

G. Thomson, The Museum Environment (Butterworth-Henemann, Stoneham, Mass., 1994).

K. V. Mardia, J. T. Kent, J. M. Bibby, Multivariate Analysis (Birnbaum and Lukacs, London, 1979).

H. Martens, T. Næs, Multivariate Calibration (Wiley, New York, 1989).

P. Geladi, H. Grahn, Multivariate Image Analysis (Wiley, New York, 1996).

R. Linari, M. Picollo, B. Radicati, “Schede di caratterizzazione di pigmenti usati in pittura,” Tech. Rep. TR/POE/92.7 [Istituto di Ricerca sulle Onde Elettromagnetiche (IROE), Firenze, Italy, July1992].

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (14)

Fig. 1
Fig. 1

(a) Schematic of the computer-controlled IS measurement setup. Quartz tungsten halogen lamps (QTH) are used as light sources; narrow-band interferential filters are lodged in the filter wheel in front of the VIS–NIR camera. (b) Our IS system during a measurement at the Uffizi Gallery, with the adapted vidicon infrared camera and the filter wheel.

Fig. 2
Fig. 2

Spectrograms of the five reflectance standards after normalization, according to a linear model of the camera transfer function.

Fig. 3
Fig. 3

Same spectrograms as are shown in Fig. 2 but now under the hypothesis of a logarithmic transfer function in the VIS region.

Fig. 4
Fig. 4

Gray-scale picture of the Signorelli’s Predella.

Fig. 5
Fig. 5

Comparison among the spectra reconstructed from the IS data (open circles) and the FORS spectra (solid curves) collected within the circled areas shown in Fig. 4.

Fig. 6
Fig. 6

Comparison between (a) the single-wavelength image (λ = 1300 nm) and (b) the PC1 image of the test panel. Carbon-black percentages are indicated.

Fig. 7
Fig. 7

Comparison between the 2-D histogram in the plane PC1/PC3 and the 2-D histogram in the plane defined by the pair of wavelengths λ = 1300 nm and λ = 560 nm. C, cinnabar; Y, yellow ochre; M, malachite; Cr, chromium oxide. The carbon-black percentages are indicated. The contour lines represent the pixel-density values (from outside to inside): 5, 20, 50, 100, 190 (full scale = 630).

Fig. 8
Fig. 8

PC1, PC2, and PC3 eigenvectors from the PCA on 29 images (taken at 420–1550 nm) of the test panel. The chromium oxide spectrum is also reported on the PC2 eigenvector plot.

Fig. 9
Fig. 9

Mean spectra (circles) of the four clusters (cinnabar, malachite, yellow ochre, and chromium oxide) corresponding to the pure pigments in the test panel (the vertical bars indicate the standards deviations). The curves represent the spectra of the same pigments as in the test panel recorded by means of a fiber-optic spectrophotometer.

Fig. 10
Fig. 10

First four PC (PC1–PC4) images from the PCA on the 29 multispectral images (taken at 420–1550 nm) of the Holy Trinity Predella by Luca Signorelli.

Fig. 11
Fig. 11

PC1 eigenvector from PCA on the 29 images of Signorelli’s predella.

Fig. 12
Fig. 12

Reconstructed reflectance spectra of the two sides of the pilaster: (a) Left side. (b) Right side. The vertical bars represent the standard deviation.

Fig. 13
Fig. 13

Score plot PC2/PC4 from the PCA on 29 monochromatic images (taken at 420–1550 nm) of Signorelli’s Predella. Cluster A corresponds to the pixels coming from the robe of the first man from the left. Cluster B corresponds to the pixels coming from the upper right-hand part of the image representing the sky.

Fig. 14
Fig. 14

Mean spectra of the pixels belonging to the clusters marked in Fig. 13. The vertical bars represent the standard deviation.

Tables (2)

Tables Icon

Table 1 Variance and Cumulative Variance for the First Three PC’s from 29 Images Taken at λ = 420–1550 nm of the Test Panel

Tables Icon

Table 2 Contributions of the Bands λ = 1300 nm and λ = 560 nm to the PC1 and PC3 Images from PCA on 29 Images Taken at λ = 420–1550 nm of the Panel

Equations (3)

Equations on this page are rendered with MathJax. Learn more.

VU j = λ j U j ,
C = DU ,
i = 1 ν   U ij 2 = 1 .

Metrics