|
| Titre : |
Machine Learning in Forestry using Remote Sensing Data study case : Classification of Forest Species |
| Type de document : |
texte imprimé |
| Auteurs : |
Yousra Belayat, Auteur ; Yasmina Saker, Auteur ; Nabila Chergui, Directeur de thèse |
| Année de publication : |
2022 |
| Importance : |
1 vol (35 f .) |
| Format : |
29cm |
| Langues : |
Français (fre) |
| Catégories : |
Thèses & Mémoires:Informatique
|
| Mots-clés : |
Forest species classification
Machine Learning |
| Index. décimale : |
004 Informatique |
| Résumé : |
The importance of forests cannot be understated. From the timber we use for building
to the air we breathe, forests are essential to our capacity to thrive.
Machine learning has been extensively used in forestry for many issues where forest
tree species classification is one of the most significant applications. It aims to categorise
trees into groups; genera or families. Besides, this task can widely benefit from
the emergence of Remote Sensing (RS) images that can remotely capture the status of
forests and facilitate their management. The integration of RS with ML can take the
classification of forest tree species to another level of precision and easiness and allow
the extraction of new insights.
This thesis aimed at classifying forest tree species based on remote sensing images
acquired from Sentinel-2 satellite and Vegetation Indices (VIs). We first extracted three
VIs, the Normalised Difference Vegetation Index (NDVI), the Enhanced Vegetation Index
(EVI) and the Atmospherically Resistant Vegetation Index (ARVI). Next, we selected
three major Machine Learning (ML) classification algorithms to perform the classification,
Decision Tree (DT), RandomForest (RF) and Support Vector Machine (SVM).
Then, we compared their performances based on four evaluation metrics; precision, recall,
F-score and accuracy. As results, the RF outperformed the other algorithms with
an accuracy of 0.76, precision = 0.71, Recall = 0.72 and F-score = 0.71. |
| Côte titre : |
MAI/0681 |
| En ligne : |
https://drive.google.com/file/d/10L8qXxREIWDvneI9KO7vPytTtjrsDGFL/view?usp=share [...] |
| Format de la ressource électronique : |
pdf |
Machine Learning in Forestry using Remote Sensing Data study case : Classification of Forest Species [texte imprimé] / Yousra Belayat, Auteur ; Yasmina Saker, Auteur ; Nabila Chergui, Directeur de thèse . - 2022 . - 1 vol (35 f .) ; 29cm. Langues : Français ( fre)
| Catégories : |
Thèses & Mémoires:Informatique
|
| Mots-clés : |
Forest species classification
Machine Learning |
| Index. décimale : |
004 Informatique |
| Résumé : |
The importance of forests cannot be understated. From the timber we use for building
to the air we breathe, forests are essential to our capacity to thrive.
Machine learning has been extensively used in forestry for many issues where forest
tree species classification is one of the most significant applications. It aims to categorise
trees into groups; genera or families. Besides, this task can widely benefit from
the emergence of Remote Sensing (RS) images that can remotely capture the status of
forests and facilitate their management. The integration of RS with ML can take the
classification of forest tree species to another level of precision and easiness and allow
the extraction of new insights.
This thesis aimed at classifying forest tree species based on remote sensing images
acquired from Sentinel-2 satellite and Vegetation Indices (VIs). We first extracted three
VIs, the Normalised Difference Vegetation Index (NDVI), the Enhanced Vegetation Index
(EVI) and the Atmospherically Resistant Vegetation Index (ARVI). Next, we selected
three major Machine Learning (ML) classification algorithms to perform the classification,
Decision Tree (DT), RandomForest (RF) and Support Vector Machine (SVM).
Then, we compared their performances based on four evaluation metrics; precision, recall,
F-score and accuracy. As results, the RF outperformed the other algorithms with
an accuracy of 0.76, precision = 0.71, Recall = 0.72 and F-score = 0.71. |
| Côte titre : |
MAI/0681 |
| En ligne : |
https://drive.google.com/file/d/10L8qXxREIWDvneI9KO7vPytTtjrsDGFL/view?usp=share [...] |
| Format de la ressource électronique : |
pdf |
|
Ajouter le résultat dans votre panier Affiner la recherche
