An Innovative Methodology

In order to attempt to answer the questions a well-defined, specific and exact methodology and protocol have been developed. For each coffin the study includes:

  1. high-resolution photogrammetric survey in situ and 3D photogrammetric data elaboration;
  2. autoptic comparison between textured 3D models and solid models to understand whether the pictorial layers correspond to the carving of the face;
  3. identification of landmarks for measurements on the model without texture (morphometric approach to facial proportion, measurements and grouping);
  4. database and data connection;
  5. new typology of yellow coffins.

1. Photogrammetric survey in situ

The project is focused on faces, but we enlarged the research on the external upper part of coffin lids until the lower part of the arms or, for the stola-coffins, until the end of the wesekh-collar. Photogrammetric surveys are performed using a full-frame Nikon D750 camera coupled with a Nikkor 35 mm f/1.8 lens and a polarizing filter mounted on the camera to avoid as many reflections and shiny areas (modern or ancient varnished areas on the yellow coffins) as possible. The polarizer has the potential to remove any reflections and consequently eliminate the main cause of noise during the dense matching reconstruction phase. The images are taken at the maximum resolution of 6016 x 4016 pixels. The camera was placed on a tripod and the
shot was delayed by 5 seconds to prevent vibration during the acquisition. This also permitted to use of low ISO values even in low-illuminated museum rooms.
The mean calculated GSD of the photogrammetric survey is around 0,2 mm since the pixel pitch is 6 μm and the distance of acquisition from the object was about 1 m. The acquisition geometry is circular around the object, maintaining the camera as much as possible nadiral to the coffin. The number of images for each coffin varies from 70 to 130 photos and there is a transversal overlap of about 79-80% of the images.

Two pre-calibrated L-shaped bars with geometrical patterns and some b/w, circle non-code targets have been created for the Project with the help of Alessandro Mandelli. These targets on the bars, which are always placed next to the coffin head during the acquisition process, are used to check the global accuracy of the 3D models and for the orientation in the space, giving the points precise coordinates. 

Photogrammetric elaboration uses Agisoft Metashape 1.7.5 following all the steps from the photo-alignment at high resolution to developing and exporting high-resolution orthophoto, both textured and not textured.

2. Analysis of the faces

Once the high-resolution 3D models and orthophotos were elaborated, the analyses of the faces started. The aim is to compare the 3D models with and without colour (visual appearance VS physical geometry) and identify the most significant facial landmarks to be measured. For this reason, two different layers have been drawn with the help of various software packages:

  1. the first layer identifies facial landmarks on the physical geometry (modelled/sculpted traits: the form of the face, nose, mouth, eyelids, eyebrows arch, ears – if present) seen from the top, right and left;
  2. the second layer for the painted features to be drawn (eyes, eyebrows, nose, lips/mouth, chin, ears – if painted), seen from the same projections as before.

These separate layers, using CAD software, can be overlapped in transparency on the solid model, on the orthophoto and between them. This overlapping allows a precise inspection of their correspondence with fixed points for the morphometric approach to facial proportion, measurements and subsequent grouping.

3. Morphometric approach to facial proportion, measurements and grouping

All the facial features are measured manually with AutoCAD software, to understand the proportions of the faces. The orthophotos that are not textured allowed the identification of the main landmarks that are fundamental for proportions.

Each face is inserted into a grid, which divides it into symmetrical portions.

The main horizontal and vertical lines intersect at the point of the nose so that the face is divided into approximate quarters varying for the different faces. Vertical and horizontal distances represent the variables related to four of the major facial components: eyes, nose, mouth and ears

The complexity of the data is recorded and stored in a form that will be used as a starting point for the future database that will manage not only the metric but also the material information of each coffin. The template form is divided into 4 parts with text and images/drawings linked to the painted/ visual appearance and physical geometry of the coffin.

A specific vocabulary, especially for the facial traits (illustrated with and without paint) and the variables, will be used for each sector. In this way, unique and coherent descriptions for each coffin will be undertaken, which are then useful for grouping and any possible seriations of the yellow coffins.

4. Database and data connection

The entire corpus of the selected coffins will be managed in a relational database including all the newly acquired data, which will be organized into different tables according to a relational structure. During this phase, the Faces Revealed Database will be made accessible to the Vatican Coffin Project team. The collaboration of the candidate with the VCP is a unique opportunity to combine data acquired from the most recent and ongoing research on yellow coffins in Europe. This collaboration will allow an integration of the data with results obtained through other scientific analyses and a comparison of the photogrammetry methodology with the latest technologies that the international team have been using: CT scans and X-rays, which allow an analysis of the thickness and density of the plaster as well as a different kind of construction and assembly; non-invasive imaging techniques (UV, IR, RTI) and non-destructive analysis (XRF) which helps to investigate painted surfaces and to collect data concerning the pigments and the nature of the various coloured materials including varnishes.

References

  • Mainieri, Stefania, Alessandro Mandelli, and Corinna Rossi, ‘Digital Humanities: A Holistic Approach to Classify Yellow Coffins’, ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  XLVI-2/W1, (25 February 2022): 335–41. (see HERE).
  • Mandelli, Alessandro, L. Perfetti, Fausta Fiorillo, Francesco Fassi, Corinna Rossi, and Christian Greco, ‘The Digitalization of Ancient Egyptian Coffins: A Discussion over Different Techniques for Recording Fine Details’, ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLII-2/W15 (2019), 743–50. (see HERE).

Go back to the Project

Please rotate your device

To continue using this app you need to rotate your device to landscape mode

Please use a bigger device

To continue using this app you need to use a device with a bigger screen such as a tablet or a laptop