Early Crop-Type Mapping Under Climate Anomalies

Abstract

Crop-type mapping is an important intermediate step for cost-effective crop management at the field level, as an overview of all fields with a particular crop type can be used for monitoring or yield forecasting, for instance. Our study used a data set with 2400 fields and corresponding satellite observations from the federal state of Bavaria, Germany. The study classified corn, winter wheat, winter barley, sugar beet, potato, and winter rapeseed as the main crops grown in Upper Bavaria. We additionally experimented with a rejection class "Other", which summarised further crop types. Corresponding Sentinel-2 data included the normalised difference vegetation index (NDVI) and raw bands from 2016 to 2018 for each selected field. The influence of raw bands compared to NDVI was analysed and the classification algorithms, i.e. support vector machine (SVM) and random forest (RF), were compared. The study showed that the use of an index should be critically questioned and that raw bands provided a wider spectral bandwidth, which significantly improved the mapping of crop types. The results underline the use of RF with raw bands and achieved overall accuracies (OA) of up to 92%. We also predicted crop types in an unknown year with significantly different weather conditions and several months before the end of the growing season. Thus, the influence of climate anomalies and the accuracy depending on the time of prediction were assessed. The crop types of a test site and year without labels could be determined with an OA of up to 86%. The results demonstrate the usefulness of the proof-of-concept and its readiness for use in real applications.

Multi-Year Time Series Transfer Learning: Application of Early Crop Classification

“…However, the ability to generalize over multiple years has not been thoroughly studied [20]. Marszalek et al [22] used Random Forest (RF) and Support Vector Machines (SVM), which have shown promise but are not scalable and require re-training for each additional observation with multi-year data.…”

Self-Supervised Learning – A Way to Minimize Time and Effort for Precision Agriculture?

“…The crop mapping task was evaluated on a data set which included the main crop types (corn, winter wheat, winter barley, winter rapeseed, sugar beet, and potato) in Upper Bavaria (Germany), collected for the years 2016, 2017 and 2018 (Figure 1). It is part of a larger collection, assembled and partly self-created, which included crop types and yields for different regions (Marszalek, 2021). The climatological data in Table 1 provided better insights into the various climatological conditions.…”

“…Initially, all 13 bands were used in this study. A more detailed band description with properties can be found in (GEE, 2012) or (Marszalek et al, 2020). For this purpose, Level-1C data were downloaded from Google Earth Engine (GEE) (Google, 2020), but not further processed to bottom-of-atmosphere (BOA) reflection information since the focus was the comparison of supervised learning with self-supervised learning.…”

“…Despite these challenging conditions, the results showed that SSL contributed to higher accuracies. We believe that the results will encourage further improvements in the field of precision farming, why the SSL framework and data will be published (Marszalek, 2021).…”

Harnessing Administrative Data Inventories to Create a Reliable Transnational Reference Database for Crop Type Monitoring

“…By addressing the motivating use-case of crop type classification using this reference dataset, data-driven methods will implicitly learn internal representations of vegetation dynamics, i.e., its phenology. Such models can further be used to monitor the development of vegetation stocks to spot unexpected patterns, e.g., caused by environmental influences [6], or to predict the expected yield [5]. Furthermore, the country-specific statistics compiled and harmonised in EUROCROPS allow for statistical investigation of regional distributions of crop cultivation patterns (cf.…”