Automatic Classification of Agricultural Crops based on Satellite Data Time Series and Neural Networks
Ochiq kirish: yoʻq
Annotatsiya
This paper addresses the development and investigation of a hybrid neural network architecture for automatic classification of agricultural crops based on satellite data time series. The proposed model combines convolutional neural networks (CNN) with long short-term memory (LSTM) networks to effectively capture both spatial features and temporal patterns in multi-temporal satellite imagery. The objective is to improve the accuracy and robustness of agricultural land classification through the integration of remote sensing data obtained from Sentinel-2 and Landsat-8 satellites across multiple growing seasons. The work analyzes spectral characteristics of vegetation and their seasonal variations using NDVI, EVI, and SAVI indices derived from time-series data. Experimental results demonstrate that the CNN-LSTM hybrid architecture achieves superior performance compared to conventional CNN approaches, with an accuracy of $94.3 \%$ on test data from agricultural areas in Central Asia.
Hali tarjima qilinmagan