Search Results (33)

The automotive industry is advancing toward fully automated driving, where perception systems rely on complementary sensors such as LiDAR and cameras to interpret the vehicle’s surroundings. For Level 4 and higher vehicles, redundancy is vital to prevent safety-critical failures. One way to achieve this is by using data from one sensor type to support another. While much research has focused on reconstructing LiDAR point cloud data using camera images, limited work has been conducted on the reverse process—reconstructing image data from LiDAR. This paper proposes a deep learning model, named LiDAR Generative Camera (LiGenCam), to fill this gap. The model reconstructs camera images by utilizing multimodal LiDAR data, including reflectance, ambient light, and range information. LiGenCam is developed based on the Generative Adversarial Network framework, incorporating pixel-wise loss and semantic segmentation loss to guide reconstruction, ensuring both pixel-level similarity and semantic coherence. Experiments on the DurLAR dataset demonstrate that multimodal LiDAR data enhances the realism and semantic consistency of reconstructed images, and adding segmentation loss further improves semantic consistency. Ablation studies confirm these findings. Full article

This study addresses the urgent need for the rapid, non-destructive assessment of key soybean components, including moisture, fat, and protein, using near-infrared (NIR) spectroscopy. This study provides technical and theoretical support for achieving the efficient and accurate detection of major soybean components and for the development of portable near-infrared (NIR) instruments. Thirty soybean samples from diverse sources were collected, and 360 spectral measurements were acquired using a 900–1700 nm NIR spectrometer after grinding and standardized sampling. To improve model robustness, preprocessing strategies such as standard normal variate (SNV), multiplicative scatter correction (MSC), and Savitzky–Golay derivatives were applied. Feature selection was conducted using competitive adaptive reweighted sampling (CARS), successive projections algorithm (SPA), and uninformative variable elimination (UVE), followed by model construction with partial least squares regression (PLSR), support vector regression (SVR), and random forest (RF). Comparative analysis revealed that the RF model consistently outperformed the others across most combinations. Specifically, the SPASNV + D₁–RF combination achieved an RPD of 14.7 for moisture, CARS–SNV + D₁–RF reached 5.9 for protein, and CARS–SG + D₂–RF attained 12.0 for fat, all significantly surpassing alternative methods and demonstrating a strong nonlinear learning capacity and predictive precision. These findings show that integrating optimal preprocessing and feature selection strategies can markedly enhance the predictive accuracy in NIR-based soybean analyses. The RF model offers exceptional stability and performance, providing both technical reference and theoretical support for the development of portable NIR devices and practical rapid-quality assessment systems for soybeans in industrial applications. Full article

Seismic sedimentology and sequence stratigraphy, as emerging interdisciplinary fields, demonstrate unique advantages in characterizing seismic geomorphological responses of various system tracts within the stratigraphic frameworks of rift lacustrine basins. Focusing on the Paleogene Dainan Formation in the Gaoyou Sag of the Subei Basin, eastern China, this study integrates seismic termination patterns, sedimentary cyclicity analysis, and well-to-seismic calibration to subdivide the formation into three third-order sequences containing lowstand (LST), transgressive (TST), and highstand (HST) system tracts. The distribution of five distinct sedimentary facies exhibits pronounced sub-tectonic zonations controlled by the basin’s architecture and structural evolution, with steep slope zones dominated by nearshore subaqueous fan–fan delta complexes, gentle slopes developing normal deltaic systems, and deep-semi-deep lacustrine facies with slump turbidite fans concentrated in depositional centers. Through a novel application of 90° phase adjustment, spectral decomposition, and multi-attribute fusion techniques, the relationship between seismic amplitude attributes and lithologies are established via seismic lithology calibration. Detailed sequence evolution analyses and seismic geomorphological interpretation systematically elucidate the spatio-temporal evolution of depositional systems within different system tracts in rift lacustrine basins, providing a novel methodological framework for sequence stratigraphic analysis in continental rift settings. Full article

In view of the higher requirements of modern machinery for multi-sensor information acquisition and fusion technology, this paper proposes a novel multi-fusion analytic mode decomposition (MFAMD) method to separate and demodulate fault features in signals. In low-speed and heavy-load equipment, the signals collected by multiple sensors contain unknown and unequal fault features and interference. Quaternion-based frequency domain fusion technology and analytically based modal extraction technology can offer novel approaches to processing large data sets in parallel while handling lengthy signals and high sampling rates. The trend spectrum segmentation method based on quaternions optimizes the hysteresis of the binary frequency. The experimental signal verifies that the proposed method is suitable for low-speed and heavy-load bearing faults. Full article

High-energy-density lithium batteries play a crucial role in the lightweight design of stratospheric airship systems. This paper conducts an in-depth experimental study of the equivalent circuit model of soft-pack batteries, with a focus on how parameter identification methods affect model accuracy. To this end, first-order RC, second-order RC, and third-order RC equivalent circuit models were constructed, and model parameters under different temperature and current conditions were obtained through constant-current intermittent discharge experiments. During the parameter identification process, special consideration was given to the impact of sampling time on voltage measurements and the interdependent constraints among models. Additionally, the effects of current, temperature, and SOC (state of charge) variations on ohmic resistance and polarization resistance–capacitor parameters were analyzed. The experimental results show that the root mean square error (RMSE) of battery terminal voltage calculated using parameter identification methods that account for these factors is significantly lower than when these factors are not considered. By comparing the voltage calculation accuracy and operational efficiency of the three models, the second-order RC model was determined to be the preferred choice due to its simple structure, high computational efficiency, and superior accuracy. Full article

Seven cyclic depsipeptides, including two new cyclic pentadepsipeptides avenamides A (1) and B (2), were isolated from a plant-derived fungus Fusarium avenaceum W8 by using the bioassay-guided fractionation method. The planar structures were elucidated by using comprehensive spectroscopic analyses, including 1D and 2D NMR, as well as MS/MS spectrometry. The absolute configuration of the amino acid and hydroxy acid residues was confirmed by using the advanced Marfey’s method and chiral HPLC analysis, respectively. Compounds 1–7 were evaluated for their cytotoxic activities against A549 and NCI-H1944 human lung adenocarcinoma cell lines and their antimicrobial activities against Staphylococcus aureus and Saccharomyces cerevisiae. As a result, compounds 1–4 showed moderate cytotoxicity, with IC₅₀ values of 6.52~45.20 µM. Compounds 1 and 3 exhibited significant antimicrobial activities against S. aureus and S. cerevisiae, with an MIC80 of 11.1~30.0 µg/mL. Full article

To address the computational complexity and deployment challenges of traditional convolutional neural networks in rice disease identification, this paper proposes an efficient and lightweight model: Ghost Channel Spatial Attention ShuffleNet with Mish-ReLU Adaptive Activation Function (GCA-MiRaNet). Based on ShuffleNet V2, we effectively reduced the model’s parameter count by streamlining convolutional layers, decreasing stacking depth, and optimizing output channels. Additionally, the model incorporates the Ghost Module as a replacement for traditional 1 × 1 convolutions, further reducing computational overhead. Innovatively, we introduce a Channel Spatial Attention Mechanism (CSAM) that significantly enhances feature extraction and generalization aimed at rice disease detection. Through combining the advantages of Mish and ReLU, we designed the Mish-ReLU Adaptive Activation Function (MAAF), enhancing the model’s generalization capacity and convergence speed. Through transfer learning and ElasticNet regularization, the model’s accuracy has notably improved while effectively avoiding overfitting. Sufficient experimental results indicate that GCA-MiRaNet attains a precision of 94.76% on the rice disease dataset, with a 95.38% reduction in model parameters and a compact size of only 0.4 MB. Compared to traditional models such as ResNet50 and EfficientNet V2, GCA-MiRaNet demonstrates significant advantages in overall performance, especially on embedded devices. This model not only enables efficient and accurate real-time disease monitoring but also provides a viable solution for rice field protection drones and Internet of Things management systems, advancing the process of contemporary agricultural smart management. Full article

The Yellow River Basin (YRB) is an important grain production base, and exploring the spatiotemporal heterogeneity and driving factors of soil erosion in the YRB is of great significance to the ecological environment and sustainable agricultural development. In this study, we employed the Revised Universal Soil Loss Equation (RUSLE) in conjunction with Transport-Limited Sediment Delivery (TLSD) to explore a modified RUSLE-TLSD for use assessing net water erosion. This modification was performed using sediment data, and the explanatory power of driving factors was assessed utilizing an optimal parameters-based geographical detector (OPGD). The results demonstrated that the modified RUSLE-TLSD can accurately simulate the spatiotemporal distribution of net water erosion (NSE = 0.5766; R² = 0.6708). From 2000 to 2020, the net water erosion modulus in the YRB ranged between 1.62 and 5.33 t/(ha·a). Specifically, the net water erosion modulus decreased in the YRB and the middle reaches of the YRB (MYRB), but it increased in the upper reaches of the YRB (UYRB). The erosion occurred mainly in the Loess Plateau region, while the deposition occurred mainly in the Hetao Plain and Guanzhong Plain. The Normalized Difference Vegetation Index (NDVI) and slope emerged as significant driving factors, and their interaction explained 31.36% of YRB net water erosion. In addition, the redistribution of precipitation by vegetation and the slope weakened the impact of precipitation on the spatial pattern of net water erosion. This study provides a reference, offering insights to aid in the development of soil erosion control strategies within the YRB. Full article

Key soil properties play pivotal roles in shaping crop growth and yield outcomes. Accurate point prediction and interval prediction of soil properties serve as crucial references for making informed decisions regarding fertilizer applications. Traditional soil testing methods often entail laborious and resource-intensive chemical analyses. To address this challenge, this study introduced a novel approach leveraging spectral data fusion techniques to forecast key soil properties. The initial datasets were derived from UV–visible–near-infrared (UV-Vis-NIR) spectral data and mid-infrared (MIR) spectral data, which underwent preprocessing stages involving smoothing denoising and fractional-order derivative[s] (FOD) transform techniques. After extracting the characteristic bands from both types of spectral data, three fusion strategies were developed, which were further enhanced using machine learning techniques. Among these strategies, the outer-product analysis fusion algorithm proved particularly effective in improving prediction accuracy. For point predictions, metrics such as the coefficient of determination (R²) and error metrics demonstrated significant enhancements compared to predictions based solely on single-source spectral data. Specifically, R² values increased by 0.06 to 0.41, underscoring the efficacy of the fusion approach combined with partial least squares regression (PLSR). In addition, based on the coverage width criterion to establish reliable prediction intervals for key soil properties, including soil organic matter (SOM), total nitrogen (TN), hydrolyzed nitrogen (HN), and available potassium (AK). These intervals were developed within the framework of the kernel density estimation (KDE) interval prediction model, which facilitates the quantification of uncertainty in property estimates. For available phosphorus (AP), a preliminary assessment of its concentration was also provided. By integrating advanced spectral data fusion with machine learning, this study paves the way for more informed agricultural decision making and sustainable soil management strategies. Full article

Traditional Chinese medicinal herbs have strict environmental requirements and are highly susceptible to weed damage, while conventional herbicides can adversely affect their quality. Laser weeding has emerged as an effective method for managing weeds in precious medicinal herbs. This technique allows for precise weed removal without chemical residue and protects the surrounding ecosystem. To maximize the effectiveness of this technology, accurate detection and localization of weeds in the medicinal herb fields are crucial. This paper studied seven species of weeds in the field of Dictamnus dasycarpus, a traditional Chinese medicinal herb. We propose a lightweight YOLO-Riny weed-detection algorithm and develop a YOLO-Riny-ByteTrack Multiple Object Tracking method by combining it with the ByteTrack algorithm. This approach enables accurate detection and localization of weeds in medicinal fields. The YOLO-Riny weed-detection algorithm is based on the YOLOv7-tiny network, which utilizes the FasterNet lightweight structure as the backbone, incorporates a lightweight upsampling operator, and adds structure reparameterization to the detection network for precise and rapid weed detection. The YOLO-Riny-ByteTrack Multiple Object Tracking method provides quick and accurate feedback on weed identification and location, reducing redundant weeding and saving on laser weeding costs. The experimental results indicate that (1) YOLO-Riny improves detection accuracy for Digitaria sanguinalis and Acalypha australis, ultimately amounting to 5.4% and 10%, respectively, compared to the original network. It also diminishes the model size by 2 MB and inference time by 10 ms, making it more suitable for resource-constrained edge devices. (2) YOLO-Riny-ByteTrack enhances Multiple Object Tracking accuracy by 3%, reduces ID switching by 14 times, and improves overall tracking accuracy by 3.4%. The proposed weed-detection and localization method for Dictamnus dasycarpus offers fast detection speed, high localization accuracy, and stable tracking, supporting the implementation of laser weeding during the seedling stage of Dictamnus dasycarpus. Full article

Multiple submerged breakwaters (MSBWs) are commonly used coastal protection structures due to their specific advantages over the emerged ones. Rip currents, as the inevitable natural hazard in the gaps of these constructions, are investigated numerically in the present study. A fully nonlinear mild-slope equation (NMSE) model possessing both fully nonlinear and fully dispersive properties is validated and adopted in the simulations. With four monochromatic wave conditions of different wave heights, periods and incidences representing low-energy, typical, storm and oblique waves tested, the flow patterns and the low-frequency oscillations of the rip currents are studied. For the convenience of risk assessment, the rip risk level is divided into three degrees according to the maximum rip flow speed. The effects of the configurations of the MSBWs on the rip current system as well as the rip risk level are examined, considering different breakwater widths, heights, forms, gap widths and gap numbers. Simulation results suggest that the cross-shore configurations of MSBWs influence the rip risk level by inducing different wave energy dissipations but the longshore configurations of MSBWs by changing flow field patterns. Full article

Early blight and late blight are two of the most prevalent and severe diseases affecting potato crops. Efficient and accurate grading of their severity is crucial for effective disease management. However, existing grading methods are limited to assessing the severity of each disease independently, often resulting in low recognition accuracy and slow grading processes. To address these challenges, this study proposes a novel two-stage approach for the rapid severity grading of both early blight and late blight in potato plants. In this research, two lightweight models were developed: Coformer and SegCoformer. In the initial stage, Coformer efficiently categorizes potato leaves into three classes: those afflicted by early blight, those afflicted by late blight, and healthy leaves. In the subsequent stage, SegCoformer accurately segments leaves, lesions, and backgrounds within the images obtained from the first stage. Furthermore, it assigns severity labels to the identified leaf lesions. To validate the accuracy and processing speed of the proposed methods, we conduct experimental comparisons. The experimental results indicate that Coformer achieves a classification accuracy as high as 97.86%, while SegCoformer achieves an mIoU of 88.50% for semantic segmentation. The combined accuracy of this method reaches 84%, outperforming the Sit + Unet_V accuracy by 1%. Notably, this approach achieves heightened accuracy while maintaining a faster processing speed, completing image processing in just 258.26 ms. This research methodology effectively enhances agricultural production efficiency. Full article

To improve the accuracy of gas outburst early warning, this paper proposes a gas outburst risk warning model based on XGBoost–GR–stacking. The statistic is based on gas outburst data from 26 mines and establishes a data generation model based on XGBoost. The obtained virtual datasets are analyzed through visualization analysis and ROC curve analysis with respect to the original data. If the augmented data has an ROC area under the curve of 1, it indicates good predictive performance of the augmented data. Grey correlation analysis is used to calculate the grey correlation degrees between each indicator and the “gas emission”. The indicator groups with correlation degrees greater than 0.670 are selected as the main control factor groups based on the sorting of correlation degrees. In this study, SVM, RF, XGBoost, and GBDT are selected as the original models for stacking. The original data and virtual data with correlation degrees greater than 0.670 are used as inputs for SVM, RF, XGBoost, GBDT, and stacking fusion models. The results show that the stacking fusion model has an MAE, MSE, and R2 of 0.031, 0.031, and 0.981. Comparing the actual and predicted values for each model, the stacking fusion model achieves the highest accuracy in gas outburst prediction and the best model fitting effect. Full article

Current stratospheric airships generally employ photovoltaic cycle energy systems. Accurately calculating their power generation is significant for airships’ overall design and mission planning. However, the power generation of solar arrays on stratospheric airships is challenging to model and calculate due to the dynamic nature of the airships’ flight, resulting in continuously changing radiation conditions on the curved surface of the airships. The power generated by the airship solar array was modeled herein through a combination of the flight attitude, spatial position, time, and other influencing factors. Additionally, the model was modified by considering the variation in photovoltaic conversion efficiency based on the radiation incidence angle, as well as the state of charge and power consumption of the energy storage battery pack. This study compared the measurement data of power generation in real flight tests with the calculation results of the model. The comparison showed that the results of the calculated model were highly consistent with the actual measured data. An average normalized root-mean-square error of 2.47% validated the accuracy of the newly built model. The generalizability and rapidity of the model were also tested, and the results showed that the model performed well in both metrics. Full article

Crop seedlings are similar in appearance to weeds, making crop detection extremely difficult. To solve the problem of detecting crop seedlings in complex field environments, a seedling dataset with four crops was constructed in this study. The single leaf labeling method was proposed as an alternative to conventional labeling approaches to improve the detection accuracy for dense planting crops. Second, a seedling detection network based on YOLOv5 and a transformer mechanism was proposed, and the effects of three features (query, key and value) in the transformer mechanism on the detection accuracy were explored in detail. Finally, the seedling detection network was optimized into a lightweight network. The experimental results show that application of the single leaf labeling method could improve the mAP0.5 of the model by 1.2% and effectively solve the problem of missed detection. By adding the transformer mechanism module, the mAP0.5 was improved by 1.5%, enhancing the detection capability of the model for dense and obscured targets. In the end, this study found that query features had the least impact on the transformer mechanism, and the optimized model improved the computation speed by 23 ms·frame⁻¹ on the intelligent computing platform Jetson TX2, providing a theoretical basis and technical support for real-time seedling management. Full article