The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLVI-4/W5-2021
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVI-4/W5-2021, 151–158, 2021
https://doi.org/10.5194/isprs-archives-XLVI-4-W5-2021-151-2021
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVI-4/W5-2021, 151–158, 2021
https://doi.org/10.5194/isprs-archives-XLVI-4-W5-2021-151-2021

  23 Dec 2021

23 Dec 2021

COLOURIMETRIC CALIBRATION FOR PHOTOGRAPHY, PHOTOGRAMMETRY, AND PHOTOMODELLING WITHIN ARCHITECTURAL SURVEY

L. Carnevali1, F. Lanfranchi1, L. Martelli1, and M. Martelli2 L. Carnevali et al.
  • 1Dept. of History, Representation and Restoration of Architecture, “Sapienza” University of Rome, Italy
  • 2Freelance computer programmer

Keywords: Architecture, Colour Imaging, Survey, Colourimetric calibration, Photogrammetry, Photomodelling

Abstract. In accordance with the “Declaration of Rome on architectural survey”, we can affirm that recording and interpretation of colour information in photographic surveying, in photogrammetric surveying and in photomodelling requires careful planning of Colour Imaging processes. Information acquired by an optical sensor is influenced not just by the actual photographed scene, but also by the spectral sensitivity of the sensor. We have adopted, from the field of Cultural Heritage, a method of colourimetric calibration for digital photographs and have proposed some adjustments to finalise this process for the purposes of Architectural Survey. With the use of a colourimetric target and a non-linear transformation algorithm, our Colour Imaging method statistically reconstructs colours conventionally unrecordable by a commercial camera. In addition, this method reconstructs colours as if the photographed object were exposed to a standard illuminant, assessing a colour error parameter value for each photo. By including the colourimetric target in every shot and by applying the calibration algorithm to all photographs taken, the process correlates all data sets to a single standard illuminant: regarding photomodelling, this leads to a more uniform and detailed representation of the surfaces of virtual models. We present two successful examples of application: one focused on a design object with physioplastic decoration and another regarding a circular fountain in a historic villa.