HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation
Master Thesis

Détection et reconnaissance de chiffres manuscrits à partir de plans cadastraux anciens par apprentissage profond

Abstract : The ancient cadastral maps are a source of important information for the study of the territory over time. It is with the aim of preserving this information that the Géomatique et Foncier Laboratory focuses on georeferencing, vectorization and mosaicing of ancient cadastral maps. Since 2016, the laboratory has implemented a processing chain. The objective of this TFE is to optimize the detection of parcel contours using parcel numbers. Since a parcel number is associated with a contour, detection of these numbers could improve the results. However, since documents are old, their numbers are handwritten and therefore difficult to detect with «classical» classification algorithms. That’s why deep learning seems perfect for this task.
Document type :
Master Thesis
Complete list of metadata

https://dumas.ccsd.cnrs.fr/dumas-03545848
Contributor : Cnam - Service Commun de la Documentation - Esgt Connect in order to contact the contributor
Submitted on : Thursday, January 27, 2022 - 3:05:26 PM
Last modification on : Monday, April 25, 2022 - 3:19:57 AM

File

GLENAZ Margot.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : dumas-03545848, version 1

Collections

Citation

Margot Glenaz. Détection et reconnaissance de chiffres manuscrits à partir de plans cadastraux anciens par apprentissage profond. Sciences de l'ingénieur [physics]. 2021. ⟨dumas-03545848⟩

Share

Metrics

Record views

3

Files downloads

4