Digital soil mapping has emerged as a new method to describe the spatial distribution of soils economically and efficiently. In this study, a lightweight soil organic matter (SOM) mapping method based on a deep residual network, which we call LSM-ResNet, is proposed to make accurate predictions with background covariates. ResNet not only integrates spatial background information around the observed environmental covariates, but also reduces problems such as information loss, which undermines the integrity of information and reduces prediction uncertainty. To train the model, rectified linear units, mean squared error, and adaptive momentum estimation were used as the activation function, loss/cost function, and optimizer, respectively. The method was tested with Landsat5, the meteorological data from WorldClim, and the 1602 sampling points set from Xinxiang, China. The performance of the proposed LSM-ResNet was compared to a traditional machine learning algorithm, the random forest (RF) algorithm, and a training set (80%) and a test set (20%) were created to test both models. The results showed that the LSM-ResNet (RMSE = 6.40, R² = 0.51) model outperformed the RF model in both the roots mean square error (RMSE) and coefficient of determination (R²), and the training accuracy was significantly improved compared to RF (RMSE = 6.81, R² = 0.46). The trained LSM-ResNet model was used for SOM prediction in Xinxiang, a district of plain terrain in China. The prediction maps can be deemed an accurate reflection of the spatial variability of the SOM distribution.

The microwave backscattering model is one of the most effective tools for surface soil moisture (SSM) inversion, which has strong theoretical support, but the inverse problem is difficult to solve. Advance in artificial intelligence offers possibilities to learn complex nonlinear relationships in a data-driven way, but it lacks physical mechanism. To combine the advantages of model-driven and data-driven methods, an SSM inversion approach that couples the AIEM-Oh model with Deep neural networks (DNNs) was proposed in this study. DNNs with different inputs were trained with a large number of simulation data generated from the AIEM-Oh model, thus embedding physical mechanisms in the data-driven scheme. Two field experiments at different scales were carried out to evaluate the performances of the proposed approach over bare surfaces. The effects of polarization modes and prior knowledge of surface roughness on SSM inversion were explored, and the accuracy of the approach was compared with the existing methods. The results suggest that satisfactory accuracy was obtained by the proposed approach, the RMSE between the measured and estimated values of SSM was 0.03-0.04 cm3·cm-3 with prior knowledge of soil roughness, and the RMSE was 0.08-0.10 cm3·cm-3 without the prior soil roughness information. VV polarization was more sensitive to SSM over bare surfaces than VH polarization. Moreover, the approach showed stable performance in different experimental regions. The results demonstrate the capability and reliability of the coupled approach for SSM inversion over bare surfaces.

Multi-beam sonar imaging detection technology is increasingly becoming the mainstream technology in fields such as hydraulic safety inspection and underwater target detection due to its ability to generate clearer images under low-visibility conditions. However, during the multi-beam sonar detection process, issues such as low image resolution and blurred imaging edges lead to decreased target segmentation accuracy. Traditional filtering methods for echo signals cannot effectively solve these problems. To address these challenges, this paper introduces, for the first time, a multi-beam sonar dataset against the background of simulated crack detection for dam safety. This dataset included simulated cracks detected by multi-beam sonar from various angles. The width of the cracks ranged from 3 cm to 9 cm, and the length ranged from 0.2 m to 1.5 m. In addition, this paper proposes a BS-UNet semantic segmentation algorithm. The Swin-UNet model incorporates a dual-layer routing attention mechanism to enhance the accuracy of sonar image detail segmentation. Furthermore, an online convolutional reparameterization structure was added to the output end of the model to improve the model’s capability to represent image features. Comparisons of the BS-UNet model with commonly used semantic segmentation models on the multi-beam sonar dataset consistently demonstrated the BS-UNet model’s superior performance, as it improved semantic segmentation evaluation metrics such as Precision and IoU by around 0.03 compared to the Swin-UNet model. In conclusion, BS-UNet can effectively be applied in multi-beam sonar image segmentation tasks.