3D FINGERPRINT DETECTION IN ANCIENT MUSEUM SCULPTURES FROM CT DATA

For museums it is important to trace the origins of the objects in their collection.

In particular, each object should have a record that includes its date of birth, place of birth and the creator's name. Unfortunately, throughout the centuries, records of museum objects may have been destroyed, lost, may not be correct, or may not have existed in the first place.

The aim of this project is to detect fingerprints in earthenware and pottery objects. The detected fingerprints can subsequently be used to uniquely determine the original sculptor of the object.

For this, first a CT scan of each object will be made. Robust image processing algorithms will be developed to search and extract fingerprint information from the CT data.

Facts

Type master project
Place internal
Supervisors Dr. Andrei Jalba,

Prof. Robert van Liere

date 11/2017

Description

1. Survey the bio-metrics literature for state-of-the-art 3D fingerprint detection algorithms in forensic law enforcement and security applications.

2. Develop image processing algorithms for finger detection in CT data. This includes:

  • the development of line extraction methods from CT data.
  • the development line coherency methods that classify line sets that belong to a finger.

3. Apply the developed methods to CT scans of

  • simple test case plaster objects.
  • a 500 year old sculpture created by the well known Dutch sculptor Johan Gregor van der Schardt.

The duration of the project is 6 months.