Interpreting a Semantic Segmentation Model for Coastline Detection
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866916251739095040 |
|---|---|
| author | O'Sullivan, Conor Coveney, Seamus Monteys, Xavier Dev, Soumyabrata |
| author_facet | O'Sullivan, Conor Coveney, Seamus Monteys, Xavier Dev, Soumyabrata |
| contents | We interpret a deep-learning semantic segmentation model used to classify coastline satellite images into land and water. This is to build trust in the model and gain new insight into the process of coastal water body extraction. Specifically, we seek to understand which spectral bands are important for predicting segmentation masks. This is done using a permutation importance approach. Results show that the NIR is the most important spectral band. Permuting this band lead to a decrease in accuracy of 38.12 percentage points. This is followed by Water Vapour, SWIR 1, and Blue bands with 2.58, 0.78 and 0.19 respectively. Water Vapour is not typically used in water indices and these results suggest it may be useful for water body extraction. Permuting, the Coastal Aerosol, Green, Red, RE1, RE2, RE3, RE4, and SWIR 2 bands did not decrease accuracy. This suggests they could be excluded from future model builds reducing complexity and computational requirements. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_11500 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Interpreting a Semantic Segmentation Model for Coastline Detection O'Sullivan, Conor Coveney, Seamus Monteys, Xavier Dev, Soumyabrata Machine Learning We interpret a deep-learning semantic segmentation model used to classify coastline satellite images into land and water. This is to build trust in the model and gain new insight into the process of coastal water body extraction. Specifically, we seek to understand which spectral bands are important for predicting segmentation masks. This is done using a permutation importance approach. Results show that the NIR is the most important spectral band. Permuting this band lead to a decrease in accuracy of 38.12 percentage points. This is followed by Water Vapour, SWIR 1, and Blue bands with 2.58, 0.78 and 0.19 respectively. Water Vapour is not typically used in water indices and these results suggest it may be useful for water body extraction. Permuting, the Coastal Aerosol, Green, Red, RE1, RE2, RE3, RE4, and SWIR 2 bands did not decrease accuracy. This suggests they could be excluded from future model builds reducing complexity and computational requirements. |
| title | Interpreting a Semantic Segmentation Model for Coastline Detection |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2405.11500 |