Bayesian Modelling of Multi-Year Crop Type Classification Using Deep Neural Networks and Hidden Markov Models
This work addresses the need for consistent yearly land-cover maps for modelling land evolution, which is important for agricultural monitoring and environmental management, representing an incremental improvement by integrating existing methods.
The paper tackles the problem of ensuring temporal consistency in yearly land-cover maps by proposing a novel approach that combines deep learning with Bayesian modelling, using Hidden Markov Models integrated with Transformer Encoder-based DNNs to classify multi-year satellite image time series for crop types. Validation on a dataset with 47 crop types over six years shows that HMMs enhance overall performance and F1 scores, demonstrating the effectiveness of the approach.
The temporal consistency of yearly land-cover maps is of great importance to model the evolution and change of the land cover over the years. In this paper, we focus the attention on a novel approach to classification of yearly satellite image time series (SITS) that combines deep learning with Bayesian modelling, using Hidden Markov Models (HMMs) integrated with Transformer Encoder (TE) based DNNs. The proposed approach aims to capture both i) intricate temporal correlations in yearly SITS and ii) specific patterns in multiyear crop type sequences. It leverages the cascade classification of an HMM layer built on top of the TE, discerning consistent yearly crop-type sequences. Validation on a multiyear crop type classification dataset spanning 47 crop types and six years of Sentinel-2 acquisitions demonstrates the importance of modelling temporal consistency in the predicted labels. HMMs enhance the overall performance and F1 scores, emphasising the effectiveness of the proposed approach.