Non-grain cultivation on arable land has posed an ever increasing challenge on national food security in sustainable agriculture. Particularly, the winter wheat–summer maize rotation systems dominate cereal production in the North China Plain. It is therefore required to accurately identify and constantly monitor the crop rotation patterns for the land-use transitions. However, existing remote sensing approaches remain dependent largely on single-temporal imagery during optimal phenological windows. Cloud contamination can frequently cause to capture the continuous and nonlinear dynamics of multi-season cropping systems. In this study, a time-series framework was developed to identify the crop rotation patterns from dynamic monitoring data. Continuous Change Detection and Classification (CCDC) algorithm was integrated with machine learning models using dense Sentinel-2 observations. A case study was also selected as the Hua County, Henan Province, China. The representative wheat–maize double-cropping region was characterized by spatial heterogeneity. All available Sentinel-2 Level-2A images were acquired from 2018 to 2024, and then processed on the Google Earth Engine (GEE) platform. The continuous multi-year spectral time series were constructed after image processing. A feature space was designed to characterize the crop phenology and surface conditions. Multi-spectral reflectance bands were incorporated with the six vegetation indices, including the Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), Bare Soil Index (BSI), Yellow Index, Normalized Difference Red Edge Index (NDREI), and Inverted Red-Edge Chlorophyll Index (IRECI). A systematic comparison was made on two classifications. One was an improved CCDC algorithm. The third-order harmonic regression was fitted into the pixel-level time series to explicitly capture intra-annual growth rhythms and long-term trends. Another was a conventional single-phase approach. Median composites were generated for the key phenological periods of wheat and maize. Annual rotation patterns were then inferred after seasonal overlay. Subsequently, the regression coefficients, harmonic components, amplitudes, and phase parameters from the CCDC model were input for Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Network (ANN) classifiers. The performance was evaluated using five-fold cross-validation and multiple accuracy metrics. The results demonstrate that the CCDC framework was achieved in the high accuracy, robustness and consistency to classify annual rotation patterns in the continuous crop transitions under variable atmospheric conditions. Among them, the highest performance was found in the CCDC–ANN combination, with an average overall accuracy of 91.8% and a Kappa coefficient of 0.891, which was improved by approximately 20% over the conventional approach. The superior performance of the ANN model was also obtained to learn complex nonlinear relationships in dense time-series features. Spatiotemporal analysis further revealed the substantial heterogeneity in the crop rotation patterns. Staple–non-staple rotations were concentrated in western hilly areas with the complex terrain, whereas the stable staple–staple and non-staple–staple systems were dominated in the eastern plains. Temporally, all major rotation types exhibited an “increase–decline–recovery” trajectory from 2018 to 2024. The great variation was primarily attributed to the policy adjustments, market dynamics, and environmental constraints. Overall, the CCDC–ANN framework can provide an accurate and scalable solution to map the wheat–maize rotation systems. The finding can also offer the strong potential to the regional monitoring of non-grain cropland and land use in modern agriculture.

Peanut is one of the most widely cultivated oil crops globally, with China leading in both production and consumption. As the demand for oil crops increases, ensuring stable peanut production and oil supply security has become a key agricultural goal. Peanut biomass, as a crucial parameter reflecting crop growth status, is essential for precision agriculture management and efficient resource utilization. The aboveground parts of peanut plants can be used not only as animal feed but also as a resource for bioenergy production. Therefore, comprehensive and accurate biomass estimation provides valuable references for yield prediction and resource management. Traditional biomass measurement methods are often labor-intensive and time-consuming, with spatial and temporal limitations. Recently, with the development of UAV remote sensing, especially the widespread application of hyperspectral imaging technology, crop biomass estimation has become more efficient. Hyperspectral imaging, known for its high resolution and rich spectral information, has been used for growth monitoring and yield estimation of crops such as soybean, rice, and wheat, demonstrating superior performance in predicting parameters like yield, chlorophyll, and nitrogen content, as well as in disease diagnosis. However, research on peanut remains limited, particularly regarding the spectral characteristics of different peanut varieties and their impact on biomass estimation accuracy. This study, using UAV hyperspectral imaging, investigated sensitive spectral bands and feature combinations for efficient and accurate field-scale peanut biomass estimation. An experimental field with 11 peanut varieties in Xingyang, Henan, was used as the study area. First, UAV hyperspectral images of the test field were collected and preprocessed with radiometric calibration and atmospheric correction to ensure data accuracy. Spectral reflectance data from ground sampling points were then extracted, and the first derivative of spectral reflectance and multiple vegetation indices were calculated to enhance the feature dimensions related to biomass. The variable importance in projection (VIP) method was used to select sensitive spectral bands and feature combinations closely related to biomass, effectively eliminating data redundancy and isolating highly relevant features. Using the selected features and ground-truth data, Support Vector Regression (SVR), back propagation neural network (BPNN), and random forest regression (RFR) models were constructed, and the estimation accuracy of different machine learning models was compared. Additionally, selected sensitive features were combined in multiple ways and input into the models to further improve estimation accuracy. The particle swarm optimization (PSO) algorithm was employed to optimize model hyperparameters, achieving the best model performance. Results showed that the sensitive features derived from the first derivative of spectral reflectance were highly correlated with peanut biomass, yielding better model performance than those derived from raw spectral reflectance and individual vegetation indices. The RF model combining the first derivative of spectral reflectance and vegetation indices achieved the highest estimation accuracy (R² = 0.75, RMSE = 0.08). Further improvement was achieved with the PSO-optimized RF model (PSO-RF), which resulted in an accuracy of R² = 0.80 and RMSE = 0.07. This study demonstrates the potential of combining UAV hyperspectral imaging with machine learning models for non-destructive peanut biomass estimation, providing essential theoretical and technical support for large-scale agricultural biomass monitoring. This study provides an effective method for accurate peanut biomass estimation and offers technical support for precision farmland management in the construction of smart villages.