Reflectance Transformation Imaging (RTI), February 2024, Biblioteca Capitolare di Verona

Dataset Description This Reflectance Transformation Imaging (RTI) dataset originates from the Palimpsests in Danger project. RTI is applied to palimpsest handwritten codices (manuscripts) from the Biblioteca Capitolare di Verona, which have been previously treated with chemical reagents. This dataset includes multiple high-resolution images taken under varying lighting angles with broadband illumination, natural color RGB images, and processed RTI files for interactive light manipulation, documenting the manuscripts' historical context and surface details. This enables thorough examination and preservation of the object Imaging Equipment Description/Specifications A digital mirrorless camera, coupled with a 90mm macro lens, is used alongside broadband illumination sources. Two stationary lights are set at a 45-degree angle to acquire a color reproduction picture. A ring light illuminates the object with direct light for a specified image capture. A handheld broadband illumination source is used for the RTI data capture. Color targets and small spherical objects are used to calibrate color, white balance, and produce RTI sets. The angle and position of the illumination source can be determined by documenting its reflection on the spheres, as well as by records of descriptive filenames. Camera - Mirrorless digital camera of 60megapixels with an IR and RGB Bayer pattern filter. Set at 100 ISO. Remotely controlled by computer and capturing software. This camera is mounted facing directly down onto the target object and spheres. Light sources – 2x 60W, 5500K broadband illumination Image Processing Description Software “Relight” is used to calibrate and process RTI data. Spheres are mapped to track the position and angle of light through the image sets, and processing techniques are used to generate different results. The most prominent among the techniques described below is Hemispherical Harmonics (HSH). See: “Material Classification using BRDF Slices” {fowang,prabath,sscher,davisg}@soe.ucsc.edu. PTM: Polynomial Texture Mapping 18 planes, or 9 using the LPTM where only the luminance change HSH: Hemispherical Harmonics 27 planes, (9 per channel) DMD: Discrete modal decomposition Described in Discrete Modal Decomposition for surface appearance modelling and rendering, it is not supported. PCA + RBF Variable number of planes using a sampled basis and radial basis function interpolation. Named as RBF[nplanes] such as RBF18. PCA + RBF + YCC Same algorithm as before, but using the Jpeg color space separately for each channel, resulting in a better (adjustable) chroma fidelity. YCC[nluma][nchroma][nchroma] such as YCC1044 for 10 luminance planes and 4+4 for the chroma. BILINEAR Bilinear interpolation replace radial basis function in the previous algorithms. Especially good for uneven sampling of the light directions. Data Structure Each item (page) has three sub-folders: JPG, TIF, and RTI. The JPG and TIF folders contain identical files in different formats, while the RTI folder contains files related to Reflectance Transformation Imaging (RTI). The JPG and TIF folders include a calibrated color picture produced by two broadband light sources (white light) at a 45-degree angle, a color image produced using a ring light attached to the lens of the photographic camera, and a series of pictures taken to produce RTI. The color picture is labeled with the name of the object. The ring light picture is labeled with the name of the object and a suffix _0_ring. RTI set pictures are labeled by a) the name of the object and b) _angle_sequence number (example: msXV007r_15_01, where 15 is the angle of the light -15 degrees- and 01 is the first image of the sequence). RTI sets are composed of 12 images (12 positions of an analog clock) and at least four different angles in increments of approximately ~16 degrees (15, 31, 47, 65). In some cases, additional increments and angles were used (~22, ~40, ~55). To View RTI Image Sets Run RelightLab application. Go to View - > View RTI. Navigate and select the folder of an RTI image set. Folder must contain info.json file. A browser window will open. Controls can be found at the lower right corner of the new RTI window. To view other RTI results navigate directly into the RTI folder of each object and open an RTI set. Open files called "Planes" (*.jpg format). Especially look for Plane6.jpg when available. Useful URLs https://www.loc.gov/preservation/digital/formats/fdd/fdd000486.shtml https://github.com/cnr-isti-vclab/relight Date of Captured Data February 2nd, 2024 Data Captured by Damianos Kasotakis