Contributor details
Affiliation
University Milano-Bicocca (Italy), University Jean
Monet (France)
University Milano-Bicocca (Italy), University Jean
Monet (France) Contributors Simone Bianco, Raimondo
Schettini, Alain Tremeau Simone Bianco, Raimondo Schettini, Alain Tremeau Contact details Spectral imaging
device details Spectral region Vis [400-700nm] Vis [400-700nm]
No. of bands Vis 31 Vis 31
File format HDF5 HDF5
Is ENVI compatible? n.a. n.a.
Storage format for hyperspectral
(BIL/BIP/BSQ) n.a. n.a.
Number of bits 16 unsigned 16 unsigned
Compression method N N
Separate header file?
(Y/N) N N
Make and model The Imaging Source DMK41AU02.AS
monochrome camera Canon 1000D 31 narrowband optical
filters
coupled with a yellow wideband filter for the second shot [Vis 5]
Weblink (if available)
Detector type CCD
1/2" CCD, progressive scan CMOS Spatial Sampling
(tentative/equivalent
ppi) n.a. n.a.
Spectral selection system (filters,
dispersive) filters filters
Details of any optics used in the acqusition(other
than those part of the
camera)
Nikon AF Nikkor 24mm f/2.8 lens with C-mount
adapter
Canon EF-S 60mm f/2.8 Macro lens
Operation since n.a. n.a.
Capture settings (Exposure time, averaging etc)
different for each filter.
from 5s to 1/5
1/160s for the first shot (unfiltered), 1/60s for the
second one (filtered) Original purpose of
the imaging system home-made system home-made system
Last calibration date 18-Mar-15 18-Mar-15
Actual Spatial
resolution n.a. n.a.
Actual Spectral
resolution Spectral Resol: ~ 10 nm
FWHM n.a.
Portability fully
equipped device transportable transportable
max dimension fully equipped device
approx. 5 cm x 5 cm x 5 cm (wide/high/large) for the
camera body
approx. 12 cm x 9 cm x 6 cm (wide/high/large) for the
camera body Details of
illuminant Type 2x ~3200K 500-Watt 2x ~3200K 500-Watt
Years of operation /
Hours of activity n.a. n.a.
Camera UM & UJM monochrome UM & UJM Canon 1000D Detatils of illuminant
spectrum grey body radiator grey body radiator Other details
Acquisition
details Aquisition date 18-Mar-15 18-Mar-15
Completion of all targets
(Fully/Partially) Full Full
Number of lights Distance btw targets
and lights distance not meaningful; distance not meaningful;
Position of the analysed target
(vertical/horizontal) 45° on a stand 45° on a stand Assessment of the
radiation impact on
the object surface n.a. n.a.
Working distance btw target and the
system 100 cm btw target and lens 130 cm btw target and lens There/which is the
max size of target to
be investigated? no no
Angle btw targets
and lights approx 2x45°/0 approx 2x45°/0
Filters /Polarisers
used? N N
Geometrical distortions (either
from camera or
setup) N (within sampling
distance) N (within sampling distance) Processing
details Processing software home made matlab scripts home made matlab scripts Recommended
Standard linear reflectance calibration on 99%
reflectance spectralon plus non-uniformity correction with neutral gray card
Standard linear reflectance calibration on 99%
reflectance spectralon plus non-uniformity correction with neutral gray card Post-processing computed using the acquisition of the MCC provided
University of Basel (Imaging & Media Lab, Switzerland)
Contributor
details Affiliation University of Basel, Digital Humanities Lab Basel, Switzerland
Contributors Giorgio Trumpy
Contact details gtrumpy@gmail.com
Spectral imaging
device details Spectral region VIS [420 - 660 nm]
No. of bands 13
File format 16-bit tif
Is ENVI compatible? No
Storage format for hyperspectral
(BIL/BIP/BSQ) n.a
Number of bits 14
Compression method none
Separate header file? (Y/N) Y
Make and model DSLR camera (Nikon D3) with lens AF-S NIKKOR 50mm f/1.4
12 Mpixel Weblink (if available)
Detector type CMOS
Spatial Sampling
(tentative/equivalent ppi) 246ppi
Spectral selection system (filters,
dispersive) Interference filters
Details of any optics used in the acqusition(other than
those part of the camera) n.a.
Operation since Capture settings
(Exposure time, averaging etc) f:8, 1/250sec Original purpose of
the imaging system Spectral analysis of artworks Last calibration date
Actual Spatial resolution 9.928 px per mm (obtained from the ruler on the image)
Actual Spectral resolution 20nm
Portability fully equipped device Yes
max dimension fully ecquipped device
Any other additional information Details of
illuminant Type Two Broncolor flashlights equipped with
umbrella diffusers Years of operation / Hours of activity
Detatils of illuminant spectrum Other details Acquisition
details Aquisition date
Completion of all targets
(Fully/Partially) Fully
Number of lights 2
Distance btw targets and lights 2X45° geometry at 50 cm distance from the object plane
Position of the analysed target
(vertical/horizontal) Horizontal
Assessment of the radiation impact on the object surface Working distance btw target and the
system
There/which is the max size of target to be investigated?
Angle btw targets and lights 2X45° geometry
Filters /Polarisers used? No
Science & Technology in Archaeology Research Centre (STARC, Cyprus)
Contributor
details Affiliation Science and Technology in Archaeology Research Centre (STARC), The Cyprus Institute, Nicosia (CY)
Contributors Vera Moitinho
Contact details veramoitinho@gmail.com
Spectral imaging
device details Spectral region VIS-NIR [420-1000 nm]
No. of bands 30
File format .CUB (cube file format)
Is ENVI compatible? YES
Storage format for hyperspectral (BIL/BIP/BSQ)
BSQ
Number of bits 8 bits
Compression method none
Separate header file?
(Y/N) Y
Make and model MuSIS HS imaging system.
Image size 1600x1200 pixels Weblink (if available) http://musis.forth-photonics.com
Detector type 1/3 inch square pixel progressive scan CCD sensor and IR to visible imaging converter (photocathode) Spatial Sampling
(tentative/equivalent ppi) Spectral selection system
(filters, dispersive)
A tuneable filtering system, integrated to the detector, enables the digital acquisition of several spectral images, which are real-time displayed on a
computer monitor.
Details of any optics used in the acqusition(other
than
those part of the camera)
n.a.
Operation since n.a
Capture settings (Exposure time, averaging
etc)
Shutter Speed (Integration Time): 1/100.000s ~ 2.0s
Original purpose of
the imaging system Hyperspectral imaging of art objects
Last calibration date n.a.
Actual Spatial resolution 5.2px per mm (obtained from the ruler on the image)
Actual Spectral resolution 20nm
Portability fully equipped
device transportable
Camera Basel, NIKON D3
Geometrical distortions (either from camera or setup) Processing
details Processing software RAW→TIF conversion with DCRAW
Recommended software for
visualisation Matlab
Reconstruction matrix (Y/N) N
Calibration workflow
(Capture to Reflectance) See separate report Post-processing details
(Distortion correction, Dark current correction, Tone curve correction,
etc.)
Dark current correction
Description of related algorithms N
Camera STARC MuSIS HS system max dimension fully
ecquipped device 139 (W) x 102 (H) x 220 (D) for the camera body Any other additional
information
Imaging monochromator and optical filtering. 34 selectable spectral bands in the range of 360-1550
nm for B & W imaging, plus color imaging mode.
Details of
illuminant Type 2 Halogen lights (EIKO: 12 Volt / 50 watt / 4700K;
Q50MR16 CG/47/36, 36 Flood, MR16 Halogen Display Lamp - Natural Daylight), set at 45 (angle).
Years of operation / Hours of activity Detatils of illuminant
spectrum Halogen lamps
Other details Acquisition
details Aquisition date Completion of all targets
(Fully/Partially) Fully
Number of lights 2
Distance btw targets and lights Position of the analysed
target (vertical/horizontal)
Horizontal
Assessment of the radiation impact on the
object surface Working distance btw target and the system There/which is the max
size of target to be investigated?
45 degree
Angle btw targets and lights
Filters /Polarisers used? There is a focus shift for data as the camera was manually focused between filters.
Geometrical distortions (either from camera or
setup)
MuSIS HS software is employed for the control of camera sensitivity, filter control and image acquisition etc., as well as for the image analysis and
data handling.
Processing
details Processing software MATLAB / ENVI
Recommended software
for visualisation N
Reconstruction matrix (Y/N)
The STARC data includes 4 versions. The camera was calibrated and two sets of data acquired and then the camera was recalibrated (in the same way as the first
calibration) and another two sets of data were acquired. This was done in order to compare the
results between calibrations and to be sure the calibration would be accurate.
Calibration workflow
(Capture to Reflectance) Dark current correction Post-processing
details(Distortion correction, Dark current correction,
Tone curve correction, etc.)
N
Centre for Sensors, Instruments and Systems Development (CD6) Technical University of Catalonia (UPC) (Spain).
Contributor details
Affiliation
Centre for Sensors, Instruments and Systems
Development (CD6) Technical University of
Catalonia (UPC)
Centre for Sensors, Instruments and Systems
Development (CD6) Technical University of
Catalonia (UPC) Contributors
Meritxell Vilaseca, Francisco J.
Burgos, Xana Delpueyo, Jorge A.
Herrera-Ramírez.
Meritxell Vilaseca, Francisco J. Burgos, Xana
Delpueyo, Jorge A.
Herrera-Ramírez.
Contact details (email &
phone) email: mvilasec@oo.upc.edu Tel.: +34 93 739 8767
details Spectral region
405-910nm
File format 16-bit tif 16-bit tif
Is ENVI compatible? No No
Storage format for
hyperspectral (BIL/BIP/BSQ) No No
Number of bits 12 12
Compression method N N
Separate header file? (Y/N) N N
Make and model Camera: QICAM Fast 1394
(Qimaging) Camera: QICAM Fast 1394 (Qimaging) LCTF: VariSpec
#VIS-07-HC-20-1012 (Verivide) + IR cutoff filter Edmund Optics
#55-234
Weblink (if available) - -
Detector type CCD CCD
Spatial Sampling
(tentative/equivalent ppi) 200ppi 200ppi
Spectral selection system
(filters, dispersive) Light-Emitting Diodes (LEDs) Liquid Crystal Tuneable Filter (LCTF) Details of any optics used in
the acqusition(other than those part of the camera)
Objective: Cinegon 1.4/12mm
Series Compact Objective: Achromatic lens FA 2/3" F1.6/25mm
Operation since 2010 2013
Capture settings Average: 20 images. F#=2.8 Exposure times varies
between bands
Average: 20 images. F#=2.8 Exposure times varies
between bands (Exposure time, averaging
etc) Original purpose of the
imaging system Spectral analysis of artwork -
Last calibration date 8th October 2014 15th October 2014
Actual Spatial resolution -
-Actual Spectral resolution Varies between bands 10nm Portability fully equipped
device Yes Yes
max dimension fully
ecquipped device 40x20x40cm 120x95x90cm
Details of
illuminant Type Trough-hole LEDs VeriVide Artificial Daylight
Flourescent Years of operation / Hours
of activity 4 1
Camera LED based MSI system LCTF based MSI system Detatils of illuminant
spectrum Narrow spectra (Mean
bandwidth =25.03nm) Continuous
Other details -
-Acquisition
details Aquisition date 8th October 2014 15th October 2014
Completion of all targets
(Fully/Partially) Fully Fully
Number of lights 4 LEDs per channel
(TOTAL=60LEDs) 2 tubes
Distance btw targets and
lights 52cm 37cm
Position of the analysed
target (vertical/horizontal) Horizontal Horizontal
Assessment of the radiation
impact on the object surface -
-Working distance btw target
and the system 52cm 100cm
There/which is the max size
of target to be investigated? 27x17.5cm(aprox) 27x17.5cm(aprox) Angle btw targets and lights 0 º (target : 68º (w.r.t. the
normal)) Diffuse illumination
Filters /Polarisers used? N LCTF
Geometrical distortions (either from camera or
setup) N N
Processing
details Processing software Matlab Matlab
Recommended software for
visualisation Anyone that could read tif
images Anyone that could read tif images
Reconstruction matrix (Y/N) N N
Calibration workflow Acquistion of a white reference and a dark/background image
Acquistion of a white reference and a dark/background image (Capture to Reflectance)
Post-processing details(Distortion correction, Dark current
correction, Tone curve correction, etc.)
Dark current correction Dark current correction
Description of related
algorithms
-Matlab (or any) script available for processing
(Y/N) N
Noise reduction methods Image average + White & Dark
current correction Image average + White &
Dark current correction
Appendix B:
Some of the Matlab functions usedMatlab functions used to read datasets:
(a) Listed each Dataset under a specific directory in MATLAB and made readable using MATLAB function.
dir : directory_name lists the files in a directory.
D = dir('directory_name') returns the results in an M-by-1 structure with the fields:
name -- Filename date -- Modification date
bytes -- Number of bytes allocated to the file isdir -- 1 if name is a directory and 0 if not
datenum -- Modification date as a MATLAB serial date number.
This value is locale-dependent.
imread: imread Read image from graphics file.
A = imread(FILENAME,FMT) reads a grayscale or color image from the file specified by the string FILENAME. FILENAME must be in the current directory, in a directory on the MATLAB path, or include a full or relative path to a file. The text string FMT specifies the format of the file by its standard file extension. The return value A is an array containing the image data. If the file contains a grayscale image, A is an M-by-N array. If the file contains a truecolor image, A is an M-by-N-by-3 array.
hdf5read: hdf5read Reads data from HDF5 files. hdf5read is not
recommended. Use H5READ instead hdf5read reads data from a data set in an HDF5 file. If the name of the data set is known, then hdf5read will search the file for the data. Otherwise, use HDF5INFO to obtain a structure describing the contents of the file. The fields of the structure returned by HDF5INFO are structures describing the data sets contained in the file. A structure describing a data set may be extracted and passed directly to hdf5read. These options are described in detail below.
DATA = hdf5read(FILENAME,DATASETNAME) returns in the variable DATA all data from the file FILENAME for the data set named
DATASETNAME.
DATA = hdf5read(FILENAME,LOCATION,ATTRIBUTENAME) returns in the variable DATA all data from the file FILENAME for the attribute named ATTRIBUTENAME attached to the location provided in LOCATION. Location can be either a dataset or a group.
multibandread :multibandread Read band interleaved data from a binary file X = multibandread
(FILENAME,SIZE,PRECISION,OFFSET,INTERLEAVE,BYTEORDER) reads band-sequential (BSQ), band-interleaved-by-line (BIL), or
band-interleaved-by-pixel (BIP) data from a binary file, FILENAME. X is a 2-D array if only one band is read, otherwise it is 3-D. X is returned as an array of data type double by default. Use the PRECISION argument to map the data to a different data type.
Matlab function to extract 24 color patches: get_CC.m
function reflectances = get_CC(CC, sampleRegion, rects )
%CC = spectral reflectance images
%sampleRegion = length in pixels of side of the square sampling region
sampleRegion = round((sampleRegion/2)-1);
[Point_x, Point_y] =
meshgrid(rects(1):floor(rects(3)/5):rects(1)+rects(3),rects(2):
floor(rects(4)/3):rects(2)+rects(4));
hold on; plot(Point_x,Point_y,'s','Color',[0 0.4470 0.7410],'MarkerSize',sampleRegion,'LineWidth',2);
hold off;
reflectances = [];
for i = 1:4 for j = 1:6
r =
mean(mean(CC(Point_y(i,j)- sampleRegion:Point_y(i,j)+sampleRegion,Point_x(i,j)-sampleRegion:Point_x(i,j)+sampleRegion,:),1),2);
reflectances = [reflectances; permute(r,[1 3 2])];
end end end
Appendix C:
Addition results from CIELAB metrics under D65 2 degree standard observerFigure 1: Color difference observed for all 24 patches from all imaging systems
Figure 2: Lightness difference observed for all 24 patches from all imaging systems 0
2 4 6 8 10 12 14 16 18 20 22
Dark skin Light skin Blue sky Foliage Blue *lower Bluish green Orange Purplish blue Moderate red Purple Yellow green Orange yellow Blue Green Red Yellow Magenta Cyan White Neutral 8 Neutral 6.5 Neutral 5 Neutral 3.5 Black
Color difference
Nikon D200 U I monchrome Canon1000D DMKmono Nikon D3 MuSIS LED LCTF
-4.0000 -2.0000 0.0000 2.0000 4.0000 6.0000 8.0000 10.0000 12.0000 14.0000
Lightness difference
Nikon D200 U I monchrome Canon1000D DMKmono Nikon D3 MuSIS LED LCTF
Figure 3: Chroma difference observed for all 24 patches from all imaging systems
Figure 4: Hue difference observed for all 24 patches from all imaging systems -20
-15 -10 -5 0 5 10 15 20
Dark skin Light skin Blue sky Foliage Blue *lower Bluish green Orange
Purplish blue Moderate red
Purple Yellow green Orange yellow Blue Green Red Yellow Magenta Cyan White Neutral 8 Neutral 6.5 Neutral 5 Neutral 3.5 Black
Hue difference
Nikon D200 U I monchrome Canon1000D DMKmono Nikon D3 MuSIS LED LCTF -15
-10 -5 0 5 10 15 20 25
Dark skin Light skin Blue sky Foliage Blue *lower Bluish green Orange Purplish blue Moderate red Purple Yellow green Orange yellow Blue Green Red Yellow Magenta Cyan White Neutral 8 Neutral 6.5 Neutral 5 Neutral 3.5 Black
Chroma difference
Nikon D200 U I monchrome Canon1000D DMKmono Nikon D3 MuSIS LED LCTF
Appendix D:
Additional RRT performed by 3 laboratoriesSource 1: Cultural heritage lab, University College London RRT performed by SEAHA student Hend Maghoub
System used: Pushbroom Line Scanner Cameras: VNIR 400-950 nm SWIR 950-2500 nm
Light source: Line of halogen lights (half bar or full bar; 500W or 250W) Calibration can be done before or after or together and then average
Maximum frame rate: 100 fps.
Lens used: VNIR: 23 mm, 30mm, 56mm SWIR: 30mm, 50 or 56mm
Datasets not yet received
Source 2: Bodleian library, University of Oxford RRT performed by Richard Mulholland
System used: Headwall Hyperspec III system Wavelength range: [380.468 – 1001.14 nm]
Total number of band: 346
Datasets collected for COSCH WG1 but did not include in the thesis due to the fact that this device is a hyperspectral system and all of the data that I received were from multispectral devices. In addition, the datasets were a large file size and processing time had increased considerably. However, being able to see in real RRT being imaged
in this lab was truly invaluable learning experience.
Source 3: Digitised suite, UCL (Centre for Digital Humanities) RRT performed by Dr. Kathryn Piquette
System used: Phase one MSI system
LED lights 15-17 shot series together with filter wheel UV protection eyeware used
No live view on screen so difficult to focus Wavelength range: [370 - 940 nm]
Total number of band: 12
RRT 1 and RRT 2 datasets were collected for COSCH WG1 but did not include in the thesis because of time factor as these datasets took more computational hours due to
large file size.