Loading...

Crop Classification using Sentinel-Image Timeseries and Deep Learning

Ghafourian Akbarzadeh, Mahnoosh | 2022

145 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 55954 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Manzuri, Mohammad Taghi
  7. Abstract:
  8. Crop classification is one of the most important applications of remote sensing in agriculture. Knowing what crops are on the farm is invaluable both on a micro and macro scale. For example, this information can be used to design and imple- ment agricultural policies, product management and ensure food security. Also, this information can be used as a prerequisite for implementing other programs at the farm scale, such as monitoring and detecting anomalies during the crop growth cycle. Most of the studies in this field are focused on the optical data of the Sentinel-2 satel- lite, but the optical data are vulnerable to atmospheric conditions, and on the other hand, there is valuable information such as a complex representation of plant crown, surface roughness, soil moisture and topography in the radar data of the Sentinel-1 satellite. In this thesis, with the intermediate fusion of the time series data of Sentinel-1 and Sentinel-2 satellites in the form of a self-supervised learning method, the accu- racy reached 98% and the F1-score reached to 0.97 and we showed that the addition of radar data improves the performance of the model, especially against cloud ob- struction, and we also showed that the model is resistant to small and imbalanced datasets, which are always one of the main challenges of agricultural data
  9. Keywords:
  10. Self-Supervised Learning ; Time Series Data ; Data Fusion ; Sentinel Satellite ; Agricultural Products ; Crop Classification

 Digital Object List

 Bookmark

...see more