A carregar...
Publicação
Deep learning for studying urban water bodies´ spatio-temporal transformation: a study of Chittagong City, Bangladesh
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2.96 MB | Adobe PDF |
Autores
Resumo(s)
Water has been playing a key role in human life since the dawn of civilization. It is an
integral part of our lives. In recent years, water bodies specially, urban water bodies
are in a poor state due to climate change and rapid urban expansion. Though some
cities have become aware of this poor state of water bodies, many cities around the
world are not contemplating this issue. Because less research has been conducted on
water bodies than other land covers in urban areas like built-up. Besides, many
advanced algorithms are currently being utilized in different fields, but in terms of
water body study, these advancements are still missing. That is why this study aims at
investigating the spatio-temporal changes in urban water bodies in Chittagong city
using deep learning and freely available Landsat data. Looking at the significance of
the study, firstly, as this study has adopted two different deep learning (DL) models
and evaluated the performance, the findings can help to understand the suitability of
applying deep learning algorithms to extract information from mid to low resolution
imagery like Landsat. Secondly, this work will help us to understand why the
conservation of the existing water bodies is so important. Finally, this study will
encourage further research in the field of deep learning and water bodies by opening
the door for monitoring other environmental resources.
Descrição
Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies
Palavras-chave
Artificial Neural Network Convolution Neural Network Deep Learning Landsat data Machine Learning Urban Water bodies