On classification of Sentinel-2 satellite images by a neural network ResNet-50
Various combinations of neural network parameters and sets of input data for satellite image classification are considered in the article. The training set is completed with a NDVI (normalized difference vegetation index) and local binary patterns. Testing of classifiers created on a different numbe...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Samara National Research University
2023-06-01
|
| Series: | Компьютерная оптика |
| Subjects: | |
| Online Access: | https://computeroptics.ru/eng/KO/Annot/KO47-3/470316e.html |
| _version_ | 1797360672958316544 |
|---|---|
| author | I.V. Bychkov G.M. Ruzhnikov R.K. Fedorov A.K. Popova Y.V. Avramenko |
| author_facet | I.V. Bychkov G.M. Ruzhnikov R.K. Fedorov A.K. Popova Y.V. Avramenko |
| author_sort | I.V. Bychkov |
| collection | DOAJ |
| description | Various combinations of neural network parameters and sets of input data for satellite image classification are considered in the article. The training set is completed with a NDVI (normalized difference vegetation index) and local binary patterns. Testing of classifiers created on a different number of epochs and samples is carried out. Values of the neural network hyperparameters are determined that allow a classification accuracy of 0.70 and an F-measure of 0.65 to be achieved. Separation into classes with similar spectral characteristics is shown to offer low classification quality at different parameters and input data sets. Additional information is required. For example, for forests to be divided into more detailed classes, one needs to employ classifiers that use images from different seasons and vegetation periods. In addition, the training set needs to be extended to take into account various natural zones, soils, etc. |
| first_indexed | 2024-03-08T15:43:03Z |
| format | Article |
| id | doaj.art-aa5b8b39c19d474787d6b04d8a6591fa |
| institution | Directory Open Access Journal |
| issn | 0134-2452 2412-6179 |
| language | English |
| last_indexed | 2024-03-08T15:43:03Z |
| publishDate | 2023-06-01 |
| publisher | Samara National Research University |
| record_format | Article |
| series | Компьютерная оптика |
| spelling | doaj.art-aa5b8b39c19d474787d6b04d8a6591fa2024-01-09T14:39:05ZengSamara National Research UniversityКомпьютерная оптика0134-24522412-61792023-06-0147347448110.18287/2412-6179-CO-1216On classification of Sentinel-2 satellite images by a neural network ResNet-50I.V. Bychkov0G.M. Ruzhnikov1R.K. Fedorov2A.K. Popova3Y.V. Avramenko4ISDCT SB RAS – Matrosov Institute for System Dynamics and Control Theory of the Siberian Branch of the RASISDCT SB RAS – Matrosov Institute for System Dynamics and Control Theory of the Siberian Branch of the RASISDCT SB RAS – Matrosov Institute for System Dynamics and Control Theory of the Siberian Branch of the RASISDCT SB RAS – Matrosov Institute for System Dynamics and Control Theory of the Siberian Branch of the RASISDCT SB RAS – Matrosov Institute for System Dynamics and Control Theory of the Siberian Branch of the RASVarious combinations of neural network parameters and sets of input data for satellite image classification are considered in the article. The training set is completed with a NDVI (normalized difference vegetation index) and local binary patterns. Testing of classifiers created on a different number of epochs and samples is carried out. Values of the neural network hyperparameters are determined that allow a classification accuracy of 0.70 and an F-measure of 0.65 to be achieved. Separation into classes with similar spectral characteristics is shown to offer low classification quality at different parameters and input data sets. Additional information is required. For example, for forests to be divided into more detailed classes, one needs to employ classifiers that use images from different seasons and vegetation periods. In addition, the training set needs to be extended to take into account various natural zones, soils, etc.https://computeroptics.ru/eng/KO/Annot/KO47-3/470316e.htmlneural networksclassificationsentinel-2remote sensingimage processing |
| spellingShingle | I.V. Bychkov G.M. Ruzhnikov R.K. Fedorov A.K. Popova Y.V. Avramenko On classification of Sentinel-2 satellite images by a neural network ResNet-50 Компьютерная оптика neural networks classification sentinel-2 remote sensing image processing |
| title | On classification of Sentinel-2 satellite images by a neural network ResNet-50 |
| title_full | On classification of Sentinel-2 satellite images by a neural network ResNet-50 |
| title_fullStr | On classification of Sentinel-2 satellite images by a neural network ResNet-50 |
| title_full_unstemmed | On classification of Sentinel-2 satellite images by a neural network ResNet-50 |
| title_short | On classification of Sentinel-2 satellite images by a neural network ResNet-50 |
| title_sort | on classification of sentinel 2 satellite images by a neural network resnet 50 |
| topic | neural networks classification sentinel-2 remote sensing image processing |
| url | https://computeroptics.ru/eng/KO/Annot/KO47-3/470316e.html |
| work_keys_str_mv | AT ivbychkov onclassificationofsentinel2satelliteimagesbyaneuralnetworkresnet50 AT gmruzhnikov onclassificationofsentinel2satelliteimagesbyaneuralnetworkresnet50 AT rkfedorov onclassificationofsentinel2satelliteimagesbyaneuralnetworkresnet50 AT akpopova onclassificationofsentinel2satelliteimagesbyaneuralnetworkresnet50 AT yvavramenko onclassificationofsentinel2satelliteimagesbyaneuralnetworkresnet50 |