Skip to Main content Skip to Navigation
Master Thesis

Développement d’un outil de segmentation et de classification semi-automatique de pierres

Abstract : Today, the work of archaeologists to study the steps of construction and reconstruction of facades is boring because they have to draw and qualify stones of walls. In order to solve this problem, we developed a set of tools. The first part, related to edge detection, is to program, using the informatic language Python, a graphical user interface and implementing our segmentation algorithm in it. The second part, concerning the supervised classification of our automatic result, uses a succession of two tools integrated in the software ArcMap to generate automatically the classification of these stones from the segmented image. Tools aiming to be semi-automatic, the user can modify every time the produced results. However, limitations are always present and particularly concerning the segmentation algorithm which is not completely adjusted to the detection of stones on degraded walls by weather or nature or whose joints are not clean.
Document type :
Master Thesis
Complete list of metadatas

https://dumas.ccsd.cnrs.fr/dumas-01168266
Contributor : Cnam - Service Commun de la Documentation - Esgt <>
Submitted on : Thursday, June 25, 2015 - 3:07:27 PM
Last modification on : Saturday, February 8, 2020 - 12:02:15 PM
Long-term archiving on: : Friday, October 9, 2015 - 5:36:04 PM

Identifiers

  • HAL Id : dumas-01168266, version 1

Collections

Citation

Pierre-Alban Hugueny. Développement d’un outil de segmentation et de classification semi-automatique de pierres. Sciences de l'ingénieur [physics]. 2014. ⟨dumas-01168266⟩

Share

Metrics

Record views

129

Files downloads

2190