Geodesic
Segmentation of forest felling using a pair of Sentinel-2 space images in winter period
Čelâbinskij fiziko-matematičeskij žurnal, Tome 9 (2024) no. 3, pp. 523-534.

Voir la notice de l'article provenant de la source Math-Net.Ru

The paper presents a comparison of segmentation models for solving the problem of detecting forest felling in winter using a pair of Sentinel-2 satellite images. The comparison includes models based on convolutional neural networks from the segmentation models library developed for the python programming language. As data for model training we used images from 2018 to 2022 from open sources of the European Space Agency, which were taken over the territory of Khanty-Mansiysk Autonomous Okrug — Yugra. These images were preprocessed to solve the following tasks: atmospheric correction of images, bringing image pairs to a single projection, slicing images into frames. Forest felling masks were manually generated since 2015 in the space services center of the Ugra Research Institute of Information Technologies. F1-measure was used to assess the quality of models, as it is required to evaluate whether the model finds all felling areas, how accurately the model finds felling areas, as well as F1-measure allows taking into account false positives of the model. The UNet++ model showed the best result with a score of 0.847. The other models showed similar results, which indicates the similarity of these models for forest harvest segmentation tasks.
Keywords: image segmentation, forest harvesting, satellite imagery, Sentinel-2, machine learning, UNet architecture. 
@article{CHFMJ_2024_9_3_a12,
     author = {V. R. Abbazov and A. V. Mel'nikov and M. A. Rusanov and O. I. Sokolkov},
     title = {Segmentation of forest felling using a pair of {Sentinel-2} space images in winter period},
     journal = {\v{C}el\^abinskij fiziko-matemati\v{c}eskij \v{z}urnal},
     pages = {523--534},
     publisher = {mathdoc},
     volume = {9},
     number = {3},
     year = {2024},
     language = {ru},
     url = {http://geodesic.mathdoc.fr/item/CHFMJ_2024_9_3_a12/}
}
TY  - JOUR
AU  - V. R. Abbazov
AU  - A. V. Mel'nikov
AU  - M. A. Rusanov
AU  - O. I. Sokolkov
TI  - Segmentation of forest felling using a pair of Sentinel-2 space images in winter period
JO  - Čelâbinskij fiziko-matematičeskij žurnal
PY  - 2024
SP  - 523
EP  - 534
VL  - 9
IS  - 3
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/item/CHFMJ_2024_9_3_a12/
LA  - ru
ID  - CHFMJ_2024_9_3_a12
ER  - 
%0 Journal Article
%A V. R. Abbazov
%A A. V. Mel'nikov
%A M. A. Rusanov
%A O. I. Sokolkov
%T Segmentation of forest felling using a pair of Sentinel-2 space images in winter period
%J Čelâbinskij fiziko-matematičeskij žurnal
%D 2024
%P 523-534
%V 9
%N 3
%I mathdoc
%U http://geodesic.mathdoc.fr/item/CHFMJ_2024_9_3_a12/
%G ru
%F CHFMJ_2024_9_3_a12
V. R. Abbazov; A. V. Mel'nikov; M. A. Rusanov; O. I. Sokolkov. Segmentation of forest felling using a pair of Sentinel-2 space images in winter period. Čelâbinskij fiziko-matematičeskij žurnal, Tome 9 (2024) no. 3, pp. 523-534. http://geodesic.mathdoc.fr/item/CHFMJ_2024_9_3_a12/

[1] Belov A.M., Myasnikov V.V., “Atmospheric correction of hyperspectral images using an approximate solution of the transport equation”, Computer optics, 38:3 (2014), 489–493 (In Russ.)

[2] Chavez P. S., Jr., “An improved dark-object subtraction technique for atmospheric scattering correction of multispectral data”, Remote Sensing of Environment, 24:3 (1988), 459–479 | DOI

[3] Atmospheric correction by DOS method, https://gis-lab.info/qa/atcorr-dos.html?ysclid=lfkpvs0on6702807400

[4] Zhai H., Zhang H., Zhang L., Li P., “Cloud/shadow detection based on spectral indices for multi/hyperspectral optical remote sensing imagery”, ISPRS Journal of Photogrammetry and Remote Sensing, 144 (2018), 235–253 | DOI

[5] Skakun S., Weavers J., Brockmann C., et al., “Cloud Mask Intercomparison eXercise (CMIX): An evaluation of cloud masking algorithms for Landsat 8 and Sentinel-2”, Remote Sensing of Environment, 274:1 (2022), 112990 | DOI

[6] Wieland M., Li Y., Martinis S., “Multi-sensor cloud and cloud shadow segmentation with a convolutional neural network”, Remote Sensing of Environment, 230 (2019), 111203 | DOI

[7] Tarasov A.V., Operational mapping of forest cover disturbances on the basis of satellite data with high spatial and temporal resolution, Thesis, Candidate of Technical Sciences, Siberian State University of Geosystems and Technologies, Thesis, Candidate of Technical Sciences, 2021 (In Russ.)

[8] Lobo Torres D., Noa Turnes J., Juan Soto Vega P., et al., “Deforestation detection with fully convolutional networks in the Amazon forest from Landsat-8 and Sentinel-2 images”, Remote Sensing, 13:24 (2021), 5084 | DOI

[9] Zhou Z., Siddiquee M. M. R., Tajbakhsh N., Liang J., “UNet++: A nested U-Net architecture for medical image segmentation”, Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, Proceedings of 4th International Workshop, DLMIA 2018, and 8th International Workshop, ML-CDS 2018, September 20, Granada, Spain, 2018, 3–11

[10] Fan T., Wang G., Li Y., Wang H., “MA-Net: A multi-scale attention network for liver and tumor segmentation”, IEEE Access, 8 (2020), 179656–179665 | DOI

[11] Sentinel-2 MSI. Processing Levels: Level-2, https://sentinel.esa.int/web/sentinel/user-guides/sentinel-2-msi/processing-levels/level-2

[12] SNAP Supported Plugins: Sen2cor, https://step.esa.int/main/snap-supported-plugins/sen2cor/

[13] Main-Knorn M., Pflug B., Louis J., Debaecker V., Müller-Wilm U., Gascon F., “Sen2Cor for Sentinel-2”, Image and Signal Processing for Remote Sensing XXIII, 10427 (2017), 1042704

[14] Lanaras C., Bioucas-Dias J., Galliani S., et al., “Super-resolution of Sentinel-2 images: Learning a globally applicable deep neural network”, ISPRS Journal of Photogrammetry and Remote Sensing, 146 (2018), 305–319 | DOI

[15] Segmentation models: Python library with neural networks for image segmentation based on PyTorch, https://github.com/qubvel/segmentation_models.pytorch#models

[16] Segmentation Models, https://segmentation-modelspytorch.readthedocs.io/en/latest/

[17] Zeiler M. D., ADADELTA: An Adaptive Learning Rate Method, 2012, arXiv: 1212.5701 | Zbl

[18] Berman M., Rannen Triki A., Blaschko M. B., The Lovász-Softmax loss: A tractable surrogate for the optimization of the intersection-over-union measure in neural networks, 2018, arXiv: 1705.08790