During the processing and planting of soybeans, it is greatly significant that a reliable, rapid, and accurate technique is used to detect soybean varieties. Traditional chemical analysis methods of soybean variety sampling (e.g., mass spectrometry and high-performance liquid chromatography) are destructive and time-consuming. In this paper, a robust and accurate method for nondestructive soybean classification is developed through hyperspectral imaging and ensemble machine learning algorithms. Image acquisition, preprocessing, and feature selection are used to obtain different types of soybean hyperspectral features. Based on these features, one of ensemble classifiers-random subspace linear discriminant (RSLD) algorithm is used to classify soybean seeds. Compared with the linear discrimination (LD) and linear support vector machine (LSVM) methods, the results show that the RSLD algorithm in this paper is more stable and reliable. In classifying soybeans in 10, 15, 20, and 25 categories, the RSLD method achieves the highest classification accuracy. When 155 features are used to classify 15 types of soybeans, the classification accuracy of the RSLD method reaches 99.2%, while the classification accuracies of the LD and LSVM methods are only 98.6% and 69.7%, respectively. Therefore, the ensemble classification algorithm RSLD can maintain high classification accuracy when different types and different classification features are used.

Mildew of maize seeds may affect their germination rates and reduce crop quality. It is crucial to classify maize seeds efficiently and without destroying their original structure. This study aimed to establish hyperspectral datasets using hyperspectral imaging (HSI) of maize seeds with different degrees of mildew and then classify them using spectral characteristics and machine learning algorithms. Initially, the images were processed with Otus and morphological operations. Each seed’s spectral features were extracted based on its coding, its edge, region of interest (ROI), and original pixel coding. Random forest (RF) models were optimized using the sparrow search algorithm (SSA), which is incapable of escaping the local optimum; hence, it was optimized using a modified reverse sparrow search algorithm (JYSSA) strategy. This reverse strategy selects the top 10% as the elite group, allowing us to escape from local optima while simultaneously expanding the range of the sparrow search algorithm’s optimal solution. Finally, the JYSSA-RF algorithm was applied to the validation set, with 96% classification accuracy, 100% precision, and a 93% recall rate. This study provides novel ideas for future nondestructive detection of seeds and moldy seed selection by combining hyperspectral imaging and JYSSA algorithms based on optimized RF.

Wide mode SAR images have an apparent incidence angle effect. The existing incident angle normalization methods assume that the relationship between the incident angle (θ) and the backscattering coefficient (σ_PQ) does not change with the growth stage of crops, which is in conflict with the real-life situation. Therefore, the normalization results of σ_PQ based on these existing methods will affect the accuracy of object classification, target recognition, and land surface parameter inversion. Here, the change in θ-σ_PQ relationship was investigated based on time-series (April to October) σ_PQ of maize canopies in northeast China, and a dynamic method based on normalized difference vegetation index (NDVI) was developed to normalize the effect of θ on σ_PQ. Through the accuracy evaluation, the following conclusions are obtained: (1) the dependence (referring to N) of Sentinel 1 C-band σ_PQ on θ varies with maize NDVI. In addition, the value of N changed from 9.35 to 0.66 at VV polarization from bare soil to biomass peak, and from 6.26 to 0.99 at VH polarization; (2) a dynamic method was proposed to quantify the change of N based on its strong correlation with NDVI, indicated by R² of 0.82 and 0.80 for VV and VH polarization, respectively; and (3) the overall root mean square error of normalized σ_PQ based on the newly-developed dynamic method is 0.51 dB, and this accuracy outperforms the original first-order cosine method (1.37 dB) and cosine square law method (1.08 dB) by about 63% and 53% on the whole. This study provides a dynamic framework for normalizing radar backscatter coefficient, improving the retrieval accuracy of land surface parameters from radar remote sensing.

The launch of the SMOS (Soil Moisture and Ocean Salinity) and SMAP (Soil Moisture Active Passive) satellites has led to the development of a series of L-band soil moisture retrieval algorithms. In these algorithms, many input parameters (such as leaf area index and soil texture) and empirical coefficients (such as roughness coefficient (h_P, N_RP) and crop structure parameter (b_P, tt_P)) are needed to calculate surface soil moisture (SSM) from microwave brightness temperature. Many previous studies have focused on how to determine the value of these coefficients and input parameters. Nevertheless, it can be difficult to obtain their ‘real’ values with low uncertainty across large spatial scales. To avoid this problem, a passive microwave remote sensing SSM inversion algorithm based on the principle of change detection was proposed and tested using theoretical simulation and a field SSM dataset for an agricultural area in northeastern China. This algorithm was initially used to estimate SSM for radar remote sensing. First, theoretical simulation results were used to confirm the linear relationship between the change rates for SSM and surface emissivity, for both H and V polarization. This demonstrated the reliability of the change detection algorithm. Second, minimum emissivity (or the difference between maximum emissivity and minimum emissivity) was modeled with a linear relationship between vegetation water content, derived from a three-year (2016–2018) SMAP L3 SSM dataset. Third, SSM values estimated by the change detection algorithm were in good agreement with SMAP L3 SSM and field SSM, with RMSE values ranging from 0.015~0.031 cm³/cm³ and 0.038~0.051 cm³/cm³, respectively. The V polarization SSM accuracy was higher than H polarization and combined H and V polarization accuracy. The retrieved SSM error from the change detection algorithm was similar to SMAP SSM due to errors inherited from the training dataset. The SSM algorithm proposed here is simple in form, has fewer input parameters, and avoids the uncertainty of input parameters. It is very suitable for global applications and will provide a new algorithm option for SSM estimation from microwave brightness temperature.

High-spatiotemporal-resolution and accurate soil moisture (SM) data are crucial for investigating climate, hydrology, and agriculture. Existing SM products do not yet meet the demands for high spatiotemporal resolution. The objective is to develop and evaluate a retrieval framework to derive SM estimates with high spatial (100 m) and temporal (<3 days) resolution that can be used on a national scale in China. Therefore, this study integrates multi-source data, including optical remote sensing (RS) data from Sentinel-2 and Landsat-7/8/9, synthetic aperture radar (SAR) data from Sentinel-1, and auxiliary data. Four machine learning and deep learning algorithms are applied, including Random Forest Regression (RFR), Extreme Gradient Boosting (XGBoost), Long Short-Term Memory (LSTM) networks, and Ensemble Learning (EL). The integrated framework (IF) considers three feature scenarios (SC1: optical RS + auxiliary data, SC2: SAR + auxiliary data, SC3: optical RS + SAR + auxiliary data), encompassing a total of 33 features. The results are as follows: (1) The correlation coefficients (r) between auxiliary data (such as sand fraction, r = −0.48; silt fraction, r = 0.47; and evapotranspiration, r = −0.42), SAR features (such as the backscatter coefficients for VV-pol (${\sigma }_{vv}^0$), r = 0.47), and optical RS features (such as Shortwave Infrared Band 2 (SWIR2) reflectance data from Sentinel-2 and Landsat-7/8/9, r = −0.39) with observed SM are significant. This indicates that multi-source data can provide complementary information for SM monitoring. (2) Compared to XGBoost and LSTM, RFR and EL demonstrate superior overall performance and are the preferred models for SM prediction. Their R² for the training and test sets exceed 0.969 and 0.743, respectively, and their ubRMSE are below 0.022 and 0.063 m³/m³, respectively. (3) The SM prediction accuracy is highest for the scenario of optical + SAR + auxiliary data, followed by SAR + auxiliary data, and finally optical + auxiliary data. (4) With an increasing Normalized Difference Vegetation Index (NDVI) and SM values, the trained models exhibit a general decrease in prediction performance and accuracy. (5) In 2021 and 2022, without considering cloud cover, the IF theoretically achieved an SM revisit time of 1–3 days across 95.01% and 96.53% of China’s area, respectively. However, SC1 was able to achieve a revisit time of 1–3 days over 60.73% of China’s area in 2021 and 69.36% in 2022, while the area covered by SC2 and SC3 at this revisit time accounted for less than 1% of China’s total area. This study validates the effectiveness of combining multi-source RS data with auxiliary data in large-scale SM monitoring and provides new methods for improving SM retrieval accuracy and spatiotemporal coverage.

A drone-borne microwave radiometer requires a high sampling frequency and a continuous acquisition capability to detect and mitigate radio frequency interference (RFI), but existing methods cannot store such a large amount of data. In this paper, the dual polling write method (DPSM) for secure digital cards triggered by a timer under a multitask framework based on STM32 MCU is proposed to meet the requirements of continuous data storage. The card programming step was changed from a query waiting structure to a polling query flag bit structure, and time-sharing processing and parallel processing were used to simulate multithreading. The experimental results were as follows: (1) the time consumption of the whole storage procedure was reduced from 4000 microseconds to 200–400 microseconds; (2) the time consumption of the card programming step was reduced from 3000 microseconds in the first block and 1000 microseconds in the second and subsequent blocks to 17–174 microseconds and 18–71 microseconds, respectively, compared with the existing method; (3) the delay in the whole sampling cycle was reduced from 3942 microseconds to 0 microseconds. The results of this paper can meet the data storage requirements of a drone-borne microwave radiometer and be applied to the high-speed storage of other devices.