What question did this study set out to answer?

The research aims to classify crops early in the season using spectral features and machine learning methods.

April 12, 2026Technologies1 citationsOpen Access

Early Crop Type Classification Based on Seasonal Spectral Features and Machine Learning Methods

AAAinagul Alimagambetova MYMoldir Yessenova AKAssem Konyrkhanova

Key Points

The research aims to classify crops early in the season using spectral features and machine learning methods.
Utilized TimeSen2Crop dataset with 1 million pixels across 16 crops.
Analyzed spectral-phenological separability through various feature sets.
Applied classical algorithms, deep learning models, and a hybrid stacking scheme.
Calculated F-criteria to assess interclass separability during vegetative growth.
Random Forest achieved macro-F1 scores of 0.46/0.74/0.75 across different scenarios.
Transformer model outperformed other deep architectures with scores of 0.42/0.68/0.78.
The hybrid multi-layer stacking scheme reached the highest quality with scores of 0.51/0.83/0.86.
Time window selection and informative indices enhanced early-season crop monitoring accuracy.

Abstract

This paper explores the feasibility of early-season crop classification based on Sentinel-2-time series using the TimeSen2Crop dataset (≈1 million pixels, 16 crops). The aim of the study was to evaluate the spectral-phenological separability of crops during the season and compare the performance of classical tabular algorithms, deep sequence models, and a seasonally oriented hybrid stacking scheme. Based on multispectral observations, a feature set was formed from 9 optical channels and 13 vegetation indices for 30 dates. F-criteria were calculated, confirming a sharp increase in interclass separability during the active vegetative growth phase and substantiating three time series truncation scenarios (early, early + mid-season, and full season). Random Forest (macro-F1: 0.46/0.74/0.75) was used as the base tabular model. LSTM, BiLSTM, GRU, 1D-CNN, and Transformer were trained in parallel, with Transformer showing the best results among the deep architectures (0.42/0.68/0.78). The main contribution of the work is a hybrid multi-layer stacking scheme combining heterogeneous base algorithms and OOF meta-features, which provides the highest quality (0.51/0.83/0.86) in all scenarios. The obtained results confirm the effectiveness of phenology-oriented selection of time windows, informative indices, and hybrid ensemble learning for improving the accuracy of early-season crop monitoring.

KI fragen

Bookmark

View Full Paper