SONAR|HES-SO

SONAR|HES-SO

SONAR|HES-SO regroupe les travaux de bachelor et master diffusables de plusieurs écoles de la HES-SO. Consultez cette page pour le détails.

En cas de question, merci de contacter les bibliothécaires de la HES-SO : bibliotheques(at)hes-so.ch

Back to SONAR
Back
Biollaz_Benjamin_2023.pdf
Bachelor thesis

Vineyards detection in aerial images

  • Sierre : Haute Ecole de Gestion Valais, 2023

Bachelor of Science HES-SO (BSc) in Business Information Technology: Haute Ecole de Gestion Valais, 2023

machine learning
semantic segmentation
convolutional neural networks
precision agriculture
U-Net
English Context: The automatic detection of vine lines offers valuable assistance to winegrowers by automating tasks such as the spraying of pesticide using drones or the vine rows counting.
Purpose: Most research reports focus solely on the vine rows detection. While this capability fulfills simple use cases, more complex tasks could use more knowledge of the environment. For instance, the optimal flight path for the drone spraying can only be planned if the obstacles are identified.
Approach: A machine learning (ML) model pre-trained only on vine lines has been provided for the thesis. To extend the binary detection to other objects, house roofs are selected as they commonly neighbor vineyards. Three methodological approaches are covered: the training from scratch of a new ML model, the transfer learning on a ML model pre-trained for vine lines, and using two separate ML models for vine lines and other objects.
Language
  • English
Classification
Computer science and technology
Notes
  • Haute Ecole de Gestion Valais
  • Informatique de gestion - Wirtschaftsinformatik
  • hesso:hegvs
Persistent URL
https://sonar.ch/hesso/documents/326785
Statistics
Document views: 163 File downloads:
  • Biollaz_Benjamin_2023.pdf: 275