Explore the vast range of titles available.

Multispectral images (MSIs) are valuable for precision agriculture due to the extra spectral information acquired compared to natural color RGB (ncRGB) images. In this paper, we thus aim to generate high spatial MSIs through a robust, deep-learning-based reconstruction method using ncRGB images. Using the data from the agronomic research trial for maize and breeding research trial for rice, we first reproduced ncRGB images from MSIs through a rendering model, Model-True to natural color image (Model-TN), which was built using a benchmark hyperspectral image dataset. Subsequently, an MSI reconstruction model, Model-Natural color to Multispectral image (Model-NM), was trained based on prepared ncRGB (ncRGB-Con) images and MSI pairs, ensuring the model can use widely available ncRGB images as input. The integrated loss function of mean relative absolute error (MRAEloss) and spectral information divergence (SIDloss) were most effective during the building of both models, while models using the MRAEloss function were more robust towards variability between growing seasons and species. The reliability of the reconstructed MSIs was demonstrated by high coefficients of determination compared to ground truth values, using the Normalized Difference Vegetation Index (NDVI) as an example. The advantages of using “reconstructed” NDVI over Triangular Greenness Index (TGI), as calculated directly from RGB images, were illustrated by their higher capabilities in differentiating three levels of irrigation treatments on maize plants. This study emphasizes that the performance of MSI reconstruction models could benefit from an optimized loss function and the intermediate step of ncRGB image preparation. The ability of the developed models to reconstruct high-quality MSIs from low-cost ncRGB images will, in particular, promote the application for plant phenotyping in precision agriculture.

Modern power distribution network incorporates distributed generation (DG) for numerous benefits. However, the incorporation creates numerous challenges in energy management and to handle the challenges it requires advanced optimization techniques for an effective operation of the network. Unlike traditional methods such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and standard Crow Search Algorithm (CSA), which suffer from premature convergence and limited adaptability to real-time variations, Reinforcement Learning Enhanced Crow Search Algorithm (RL-CSA) which is proposed in this research work solves network reconfiguration optimization problem and minimize energy losses. Unlike conventional heuristic methods, which follow predefined search patterns, RL-CSA dynamically refines its search trajectory based on real-time feedback, ensuring superior convergence speed and global search efficiency. The novel RL-CSA enables real-time adaptability and intelligent optimization for energy loss reduction in distributed networks. The proposed model validation is performed on the IEEE 33 and 69 Bus test systems considering diverse performance metrics such as power loss reduction, voltage stability, execution time, utilization efficiency for DG deployment, and energy cost minimization. Comparative results show that RL-CSA achieves a 78% reduction in energy losses, limiting power loss to 5 kW (IEEE 33-Bus) and 8 kW (IEEE 69-Bus) whereas traditional models converge at higher loss levels. The execution time is optimized to 1.4 s (IEEE 33-Bus) and 1.8 s (IEEE 69-Bus), significantly faster than GA, PSO, and CSA, making RL-CSA more efficient for real-time power distribution applications. By balancing exploration-exploitation using CSA while adapting search parameters through reinforcement learning, RL-CSA ensures scalability, improved DG utilization (98%), and better voltage stability (< 0.005 p.u.), making it a robust and intelligent alternative for modern smart grid optimization.

The complexity of urban road scenes at night and the inadequacy of visible light imaging in such conditions pose significant challenges. To address the issues of insufficient color information, texture detail, and low spatial resolution in infrared imagery, we propose an enhanced infrared detection model called LFIR-YOLO, which is built upon the YOLOv8 architecture. The primary goal is to improve the accuracy of infrared target detection in nighttime traffic scenarios while meeting practical deployment requirements. First, to address challenges such as limited contrast and occlusion noise in infrared images, the C2f module in the high-level backbone network is augmented with a Dilation-wise Residual (DWR) module, incorporating multi-scale infrared contextual information to enhance feature extraction capabilities. Secondly, at the neck of the network, a Content-guided Attention (CGA) mechanism is applied to fuse features and re-modulate both initial and advanced features, catering to the low signal-to-noise ratio and sparse detail features characteristic of infrared images. Third, a shared convolution strategy is employed in the detection head, replacing the decoupled head strategy and utilizing shared Detail Enhancement Convolution (DEConv) and Group Norm (GN) operations to achieve lightweight yet precise improvements. Finally, loss functions, PIoU v2 and Adaptive Threshold Focal Loss (ATFL), are integrated into the model to better decouple infrared targets from the background and to enhance convergence speed. The experimental results on the FLIR and multispectral datasets show that the proposed LFIR-YOLO model achieves an improvement in detection accuracy of 4.3% and 2.6%, respectively, compared to the YOLOv8 model. Furthermore, the model demonstrates a reduction in parameters and computational complexity by 15.5% and 34%, respectively, enhancing its suitability for real-time deployment on resource-constrained edge devices.

To address the insufficient localization accuracy of wireless sensor networks (WSNs) in complex underground coal mine tunnel environments caused by signal fluctuations and dynamic node movement, this paper proposes an iterative weighted centroid localization algorithm based on Received Signal Strength Indicator (RSSI) clustering. The algorithm optimizes RSSI data using K-means clustering to dynamically acquire path loss parameters and achieves high-precision localization by integrating an improved iterative weighted centroid algorithm. The experimental data show that, compared with several currently high-performance localization algorithms, the algorithm proposed in this paper exhibits certain performance advantages in different scenarios such as adjustment of node communication radius, change of beacon node ratio, and variation of tunnel width, which improves the localization robustness in complex environments.This study provides a theoretical reference for three-dimensional localization in confined, elongated spaces such as underground mine tunnels.

This paper proposes a control scheme based on an optimal triple phase-shift (TPS) control for dual active bridge (DAB) DC–DC converters to achieve maximum efficiency. This is performed by analyzing, quantifying, and minimizing the total power losses, including the high-frequency transformer (HFT) and primary and secondary power modules of the DAB converter. To analyze the converter, three operating zones were defined according to low, medium, and rated power. To obtain the optimal TPS variables, two optimization techniques were utilized. In local optimization (LO), the offline particle swarm optimization (PSO) method was used, resulting in numerical optimums. This method was used for the low and medium power regions. The Lagrange multiplier (LM) was used for global optimization (GO), resulting in closed-form expressions for rated power. Detailed analyses and experimental results are given to verify the effectiveness of the proposed method. Additionally, obtained results are compared with the traditional single phase-shift (SPS) method, the optimized dual phase-shift (DPS) method, and TPS method with RMS current minimization to better highlight the performance of the proposed approach.

In order to reduce magnetic components for a current-fed dual active bridge converter, this paper proposes a dual coupled-inductor (DCI) structure, which integrates two DC inductors, one high-frequency transformer, and one leakage inductor into two EE cores. By analyzing the principle of the magnetic components, a derivation process is presented to modify the current-fed dual active bridge converter. To simplify the design and enhance efficiency, an equal air gap length optimization method is proposed. And the geometric parameters with the highest efficiency are optimized based on losses. Finally, the feasibility and effectiveness of the above design were verified through a 1 kW test prototype.

To address challenges related to the inadequate representation and inaccurate discrimination of pedestrian attributes, we propose a novel method for person re-identification, which leverages global feature learning and classification optimization. Specifically, this approach integrates a Normalization-based Channel Attention Module into the fundamental ResNet50 backbone, utilizing a scaling factor to prioritize and enhance key pedestrian feature information. Furthermore, dynamic activation functions are employed to adaptively modulate the parameters of ReLU based on the input convolutional feature maps, thereby bolstering the nonlinear expression capabilities of the network model. By incorporating Arcface loss into the cross-entropy loss, the supervised model is trained to learn pedestrian features that exhibit significant inter-class variance while maintaining tight intra-class coherence. The evaluation of the enhanced model on two popular datasets, Market1501 and DukeMTMC-ReID, reveals improvements in Rank-1 accuracy by 1.28% and 1.4%, respectively, along with corresponding gains in the mean average precision (mAP) of 1.93% and 1.84%. These findings indicate that the proposed model is capable of extracting more robust pedestrian features, enhancing feature discriminability, and ultimately achieving superior recognition accuracy.

The widespread use of digital image tampering has created a strong need for accurate and generalizable detection systems, especially in domains like forensics, journalism, and cybersecurity. Traditional handcrafted methods often fail to capture subtle manipulation artifacts, and many deep learning approaches lack generalization across diverse image sources and manipulation techniques. To address these limitations, we propose a tampered image classification model based on transfer learning using EfficientNetV2B0. This backbone is combined with a lightweight, regularized CNN classification head and optimized using Focal Loss to address class imbalance. The architecture integrates compound scaling, fused MBConv layers, and squeeze-and-excitation (SE) attention to improve feature representation and robustness. We evaluate the model on four benchmark datasets-CASIA v1, Columbia, MICC-F2000, and Defacto (Splicing)-and achieve exceptional performance, with AUC scores up to 1.0000 and F1-scores up to 0.9997. Comparisons with 42 state-of-the-art models, including IML-ViT, MVSS-Net++, ConvNeXtFF, and DRRU-Net, show our method consistently outperforms existing approaches in accuracy, precision, recall, and generalization, particularly on high-resolution and compressed images. These results demonstrate the practical effectiveness and forensic reliability of the proposed system.

This paper proposes a bidirectional feature fusion pyramid (B-FPN) Single Shot Multiple Frame Detector (SSD) algorithm. First, a bidirectional feature pyramid (B-FPN) structure is constructed, which realizes the bidirectional fusion of the feature layers and improves the accuracy of detection. Second, we introduce coordinate attention (CA) to focus on the important channel features while preserving their location information, thereby increasing the focus on the important information. Finally, optimizing the loss function speeds up the convergence of the model and further improves the detection accuracy of the network. The experimental results show that on the VOC2007 dataset, the mAP of the algorithm in this paper is 76.48%, which is 3.52% higher than that of the SSD algorithm. On the COCO 2017 dataset, the mAP of the proposed algorithm is 3.85% higher than that of the SSD algorithm. Compared with other mainstream target detection algorithms, the algorithm in this paper has certain advantages in detection accuracy, and can also achieve satisfactory results in detection speed. Finally, the accuracy of foreign object recognition in the special environment of iron ore transportation is 98.26%.

Power loss optimization aiming at the high-efficiency drive of front-and-rear-induction-motor-drive electric vehicle (FRIMDEV) as an effective way to improve energy efficiency and extend driving range is of high importance. Different from the traditional look-up table method of motor efficiency, power loss optimization of the dual- motor system based on the loss mechanism of induction motor (IM) is proposed. First of all, based on the power loss characteristic of FRIMDEV from battery to wheels, the torque distribution optimization model aiming at the minimum system power loss is put forward. Secondly, referring to d-q axis equivalent model of IM, the power loss functions of the dual-IM system are modeled. Then, the optimal torque distribution coefficient (β o) between the two IMs is derived, and the theoretical switching condition (T sw) between the single- and dual-motor-drive mode (SMDM and DMDM) is confirmed. Finally, a dual-motor test platform is developed. The derived torque distribution strategy is verified. The influence of motor temperature on β o and T sw are tested, and the correction models based on temperature difference are proposed. Based on the system power loss analysis, it can be confirmed that, under low load conditions, the SMDM takes priority over the DMDM, and the controller of the idling motor should be shut down to avoid the additional excitation loss. While under middle to high load conditions, even torque distribution (β o = 0.5) is preferred if the temperature difference between the two IMs is small; otherwise, β o should be corrected based on dual-motor temperatures. The theoretical T sw derived without dealing with temperature difference is a function only of motor speed, while temperature difference correction of it should be conducted in actual operations based on motor resistance changing with temperature.