Underground fires pose a significant threat to production safety in coal mines, and existing detection methods suffer from drawbacks such as poor adaptability to complex subterranean environments and excessive model parameters. To address the need for deploying object detection models on resource-constrained devices, this paper proposes a novel and efficient algorithm forUndergroundFireYOLOdetection, named UF-YOLO. The core innovation of this method is threefold: first, the StarNet module is introduced into the backbone to significantly reduce model parameters and computational complexity without sacrificing accuracy; second, the Cross-scale Context Fusion Module (CCFM) is integrated into the neck to enhance the model’s detection capability for fires of various scales, particularly small targets; and finally, Partial Convolution (PConv) is integrated to extract spatial features more efficiently, further reducing redundant computations and memory access. On our self-built Mine Fire Image Dataset (MFID), compared to the baseline model YOLOv11m, UF-YOLO reduces parameters by 77.1%, increases inference speed by 60.6%. Experimental results on the public COCO val 2017 dataset demonstrate that the proposed method outperforms state-of-the-art (SOTA) models such as YOLOv12. The results confirm that UF-YOLO can be efficiently deployed on the edge-side of coal mine IoVT monitoring systems to performe accurate and real-time fire detection. This work provides a new intelligent paradigm for the real-time monitoring of underground fires.

The security of voice control in the Voice Internet of Things (Voice IoT) heavily relies on the fast and accurate authentication of the command issuer. In this work, we focus on the critical application scenario of Voice IoT in underground coal mines, where voice commands typically last 4–10 seconds. Speech in this scenario typically consists of short, imperative utterances and faces challenges from environmental noise and device heterogeneity. The limitations of traditional speaker verification models in temporal modeling restrict their performance in such scenarios. To address this, this paper proposes a three-dimensional attention module (Branch-MFA) designed for Voice IoT. This module employs a dual-parallel branch architecture: the MFA branch is responsible for extracting attention in the frequency and channel dimensions, and its multi-scale nature enables it to effectively focus on speaker-discriminative frequency bands that remain stable under noise and different collection devices, thereby enhancing the model’s environmental robustness; the GLTA branch, through its innovative grouped variable-length attention mechanism, specifically models the temporal structure of these short voice commands, addressing the challenge of sparse temporal information in short utterances. By integrating the dual-branch outputs through a fusion module, we construct the Branch-MFA-TDNN model. Experiments on the Cn-Celeb dataset show that this model significantly outperforms baseline models in short-utterance verification tasks, particularly for the challenging 4–10 second duration relevant to mine communications, providing an identity authentication solution for Voice IoT that combines high security and real-time performance. We have also released the code1for future comparison.

The Internet of Nano Things (IoNT) refers to an interconnected network of nanoscale components engineered to perform tasks such as data processing, storage, and actuation. IoNT has broad applications, including environmental monitoring and pollution source localization. In order to achieve monitoring and localization for multiple releasing sources (RSs), the deployment of nanosensor networks is indispensable. However, constrained by spatial limitations and high costs, sensors can only be sparsely deployed, resulting in severe degradation in localization performance. In this paper, we consider a turbulent diffusion molecular communication scenario and the objective is to enable multi-source localization and obstacle perception with sparse nanosensors. For sparse signal recovery, we first propose a real-symmetric based on Truncated Nuclear Norm Regularization with Alternating Direction Method of Multipliers (RS-TNNR) matrix completion algorithm, which utilizes the spatial symmetry of molecular diffusion to achieve precise data recovery under high missing ratios. Furthermore, for multi-source localization and obstacle perception, we also propose an Adaptive Iterative Grid based on Sparse Bayesian Learning (AIG-SBL) algorithm, which enhances the localization accuracy with SBL, mitigates off-grid errors via the proposed adaptive iterative grid, and simultaneously estimates obstacle position and radii. Simulation results demonstrate the effectiveness of the proposed algorithms for RS-TNNR and AIG-SBL.

This paper addresses the challenges of automatic monitoring and restoration in ancient mural conservation, aiming to enhance the efficiency and quality of heritage preservation. Traditional manual inspection is time-consuming and often misses early damage, while existing digital restoration models struggle with consistent restoration, especially for large-scale damage. To address these issues, we propose an Internet of things (IoT)-based solution combining event-driven edge intelligence and artificial intelligence generated content (AIGC) techniques. A fine-tuned EdgeSAM model, using a Conv-adapter, enables efficient damage segmentation at the edge; an event-driven mechanism reduces resource consumption; and a LoRA-tuned PowerPaint model, aided by Blip2 and Qwen, provides effective restoration of large damaged areas. Cloud-side processing utilizes AIGC techniques to restore damaged mural areas, ensuring high-quality restoration while minimizing communication demands. Experimental results demonstrate that the proposed method achieves accurate damage monitoring on resource-constrained edge devices and generates diverse, contextually appropriate restoration results on cloud servers, providing a deployment-oriented feasibility validation under simulated temporal degradation and real hardware constraints.

Perceiving scene depth and 3D structure is fundamental for environmental comprehension and interaction in Internet of Video Things (IoVT) devices. Self-supervised monocular depth estimation, which leverages photometric consistency across sequential video frames without requiring ground-truth labels from active sensors, has emerged as a compelling paradigm. Nevertheless, this paradigm suffers from inherent limitations in low-texture or occluded regions, where photometric supervision becomes ambiguous or invalid, typically leading to structural degradation and indistinct object boundaries when inferring depth. To mitigate this limitation, we propose a dual-branch local distillation framework that harnesses priors from depth foundation models (DFMs) to alleviate depth ambiguity and improve fine-grained estimation accuracy. Specifically, guided by a frozen DFM, the framework synergistically employs local-context and cross-context supervision to optimize a student network, facilitating robust perception of depth discontinuities and precise boundary modeling. Furthermore, to enhance the spatial representation within a lightweight architecture, we design DE-LiteMono as the student model, which recovers geometric details through a detail enhancement block and fusion modules. Moreover, a spatial distance consistency loss is introduced to explicitly model the relative geometric topology in the image plane, providing robust structural guidance. Extensive experiments on standard benchmarks, including KITTI, Cityscapes, and Make3D, demonstrate that the proposed method outperforms state-of-the-art self-supervised methods, achieving sharper depth boundaries, reduced ambiguity, and superior quantitative performance.

Industrial IoVT (Internet of Video Things) still faces the dual bottleneck of insufficient accuracy and poor real-time performance in circuit board tiny defect detection. To this end, we propose RGM-YOLO (RefConv–GhostNet–CBAM-enhanced YOLOv8 ), which introduces deformable convolution and channel attention via RefConv and GhostNet modules, and experimentally validates it on the BDL-PCB (Bare Die on Laminate–Printed Circuit Board) large-scale dataset. Experimental results show that RGM-YOLO achieves 94.2% in mAP50 and 67.3% in mAP90–95, representing improvements of 2.4% and 11.2% over the baseline model, YOLOv8. The number of parameters and GFLOPs is reduced by 4.2M and 2.5G, respectively, while the FPS increases from 78 to 102. This approach offers a high-precision, low-latency defect detection paradigm for edge IoVT devices targeting small defects and can be generalized to other industrial quality-inspection scenarios.

Recently, edge computing has emerged as a prevailing paradigm in applying deep learning-based object detection models, offering a promising solution for time-sensitive tasks. However, existing edge object detection faces several challenges: 1) These methods struggle to balance detection precision and model lightweightness. 2) Existing generalized edge-deployment designs offer limited adaptability for object detection. 3) Current works lack real-world evaluation and validation. To address these challenges, we propose theEdgeDetectionToolbox(ED-TOOLBOX), which leverages generalizable plug-and-play components to enable edge-site adaptation of object detection models. Specifically, we propose a lightweightReparameterized Dynamic Convolutional Network(Rep-DConvNet) that employs a weighted multi-shape convolutional branch structure to enhance detection performance. Furthermore, ED-TOOLBOX includes aSparse Cross-Attention(SC-A) network that adopts a localized-mapping-assisted self-attention mechanism to facilitate a well-craftedJoint Modulein adaptively transferring features for further performance improvement. Moreover, we propose anEfficient Headfor the classification and location modules to achieve more efficient prediction. Additionally, in practical industrial scenarios, we identify that helmet detection-one of the most representative edge object detection tasks-overlooks band fastening, which introduces potential safety hazards. To address this, we build aHelmet Band Detection Dataset(HBDD) and apply an edge object detection model optimized by the ED-TOOLBOX to tackle this real-world task. Extensive experiments validate the effectiveness of components in ED-TOOLBOX. In visual surveillance simulations, ED-TOOLBOX-assisted edge detection models outperform sixstate-of-the-artmethods, enabling real-time and accurate detection. These results demonstrate that our approach offers a superior solution for edge object detection.

Cross-modal federated learning is constrained by bandwidth and on-device compute. We present Mobiflip: a minimalist strategy that freezes a lightweight backbone and communicates only a channel-wise \(1\times 1\) scaling adapter appended to the image branch. Guided by the Information Bottleneck, we prove that under common distributional and linear-encoder surrogates, per-channel scaling attains the linear optimum; coupled with the directional geometry of (Mobile)CLIP, the adapter is, in first-order approximation, an optimal preconditioner of the cosine-similarity space—preserving discriminative directions while compressing redundancy and suppressing inter-client drift. We adopt MobileCLIP as a mobile-friendly backbone to jointly minimize compute and communication. On CIFAR-10/100 and medical imaging, a single aggregation already yields stable Bacc; each round transmits only about 0.7% of backbone parameters with \(>\!\!92\%\) reduction in communication. Compared with recent federated multimodal/large-model methods, Mobiflip maintains—or even improves—accuracy under ultra-low communication.

The integrity of electronic health record (EHR) is susceptible to compromise by hardware failures, software errors, or human errors. To date, numerous data integrity verification schemes have been proposed, but most face challenges related to third-party auditing and communication overhead. To address this, a novel EHR integrity verification scheme based on a multi-branch authentication tree is presented in this paper. By integrating an edge-based batch processing mechanism with data identity labeling technology, a low-overhead data verification framework is constructed, effectively reducing communication load. A minimal multi-branch tree structure is innovatively designed to enable parallel authentication and batch signing of data blocks. Concurrently, a random security code generation algorithm is introduced to ensure data security. Experimental and analytical results demonstrate that the proposed scheme maintains correctness, efficiency, and security, consistently achieving 100 % precision in detecting corrupted EHR data replicas. This scheme provides an efficient and reliable data integrity guarantee mechanism for EHR within edge computing environments and contributes significantly to building a trustworthy medical service system.

In Internet of things (IoT) systems enabled by 6G, unmanned aerial vehicles (UAVs), acting as communication nodes, have the advantages of flexible deployment and wide-area coverage. The channel prediction capability of UAVs for ground communication is of great significance for improving the reliability of IoT communication systems. We propose an innovative and interpretable paradigm for channel prediction based on “physical feature extraction + machine learning”. Specificallywe proposes a real-time UAV-to-ground channel prediction method that leverages propagation environment sensing data, aiming to enhance prediction accuracy and generalization by deeply integrating environmental and communication information. Firstly, we construct the first UAV sensing-communication integrated dataset featuring multi-band, multi-dimensional channel parameters, including UAV-to-ground RGB images, depth maps, and channel data. We then extract multimodal features with clear physical significance relevant to wireless propagation, such as relative position, relative altitude, relative volume, and transmitter-receiver distance. Finally, this paper designs a fusion architecture based on convolutional neural network (CNN) and multilayer perceptron (MLP). This architecture takes multimodal feature data as input, utilizes CNN to extract local features of multi-modal features, and models the fusion of multi-modal features through MLP. Experimental results demonstrate that our model consistently outperforms comparative model. Importantly, our feature analysis quantitatively reveals—for the first time—that building volume is the most influential factor in channel behavior, and that prediction accuracy degrades with increasing flight altitude. Furthermore, system-level simulations confirm that channel prediction leads to substantial improvements in network performance. This work presents a robust and interpretable framework for environment-aware channel characterization, laying a foundation for future 6G intelligent communication systems.

With the rapid advancement of technologies such as artificial intelligence and the Internet of Things, video surveillance—being a core component of video IoT systems—has been widely adopted for underground coal mine safety monitoring. However, the dim lighting and heavy coal dust in underground mines result in poor visibility and significant detail loss in monitoring images, posing a major challenge to coal mine safety management. To address these issues, we propose a low-light image enhancement method tailored for underground coal mine environments, based on Zero-DCE. In our method, traditional convolutions are replaced with Ghost modules to reduce computational cost while maintaining feature extraction capability. Additionally, we incorporate global context blocks and a Vision Transformer branch to integrate more global information into the model. Specifically, the global context blocks improve the model’s ability to correct uneven illumination and prevent overexposure. Meanwhile, the Vision Transformer branch captures long-range dependencies and fuses local and global features to enhance brightness while mitigating color distortion. Furthermore, we replace the original quadratic iterative function with a reciprocal illumination mapping function, enabling more stable and perceptually aligned brightness adjustments. Experimental results on the coal mine underground personnel dataset demonstrate that our method outperforms several state-of-the-art low-light enhancement techniques, achieving superior results in both qualitative and quantitative evaluations. These findings indicate that our approach significantly improves the visibility and overall quality of underground coal mine monitoring images.

With the rapid development of Visual Internet of Things (VIoT) and text detection technology, they have been widely combined and applied to many industrial production sites, such as label text detection, achieving impressive results. However, there are still many shortcomings in the text detection technology: 1) the existing VIoT system has very limited detection precision for text with large scale changes, especially for some small-scale text detection; 2) the existing text detection algorithms cannot meet the actual situation, as the labels often contain handwritten texts, and the text to be detected is arbitrary shape; and 3) in the actual detection, there are many creases or defects on the text label. To solve the above problems, this article designs a text detection method based on a multiscale selection fusion feature pyramid and multisemantic spatial network (MSNet) to assist the VIoT system in detecting label text. First, a multiscale selective fusion feature pyramid is designed, which not only uses the texture extraction module to effectively improve the text texture feature and multiscale feature extraction ability, but also uses the cross-scale selective fusion block to selectively fuse the features of different stages to reduce the influence of pollution on detection. In addition, a MSNet is designed to capture the multisemantic spatial information of each feature channel by using the multiscale deep shared 1-D convolution, which effectively integrates global context dependence and multisemantic spatial prior. Experimental results show that the comprehensive index F-measure on the public datasets ICDAR2015, total-text, and CTW1500 is increased by 5.7%, 3.3%, and 3.8%, respectively. Furthermore, the precision, recall, and F-measure on the dataset label-text are 94.6%, 90.7%, and 92.6%, respectively. The label text detection VIoT system we designed has been deployed in the field and achieved excellent performance. The code of our proposed method can be found in:https://github.com/rebornone1/MSNet

The explosive growth of remote-sensing images generated by emerging space–air–ground integrated IoT networks makes centralized detector training infeasible due to limited bandwidth and strict data privacy constraints. While lightweight single-stage object detectors offer efficiency, they suffer significant accuracy degradation for small, dense, and arbitrarily oriented targets. Furthermore, existing federated object detection frameworks typically neglect client heterogeneity. To overcome these limitations, we propose a two-stage personalized federated detection framework. In Stage 1, we independently train a conventional single-stage rotated object detector on each client and aggregate model updates using an adaptive similarity momentum aggregation (ASMA) strategy, effectively pooling knowledge across non-IID client datasets to improve global generalization. In Stage 2, each client is equipped with a private selective depthwise attention convolution (SDAConv) module, leveraging Stage-1 priors to reconstruct fine-grained, client-specific features without additional communication overhead, thus tailoring predictions to local data distributions. Experiments conducted on five non-IID splits derived from DOTA-1.0, along with DIOR and VisDrone datasets, demonstrate improvements of up to +3.5 mAP compared to federated learning baselines under the same communication budget, simultaneously maintaining global robustness and enhancing local detection accuracy.

Clear monitoring images are crucial for the safe operation of coal mine Internet of Video Things (IoVT) systems. However, low illumination and uneven brightness in underground environments significantly degrade image quality, posing challenges for enhancement methods that often rely on difficult-to-obtain paired reference images. Additionally, there is a tradeoff between enhancement performance and computational efficiency on edge devices within IoVT systems.To address these issues, we propose a multimodal image enhancement method tailored for coal mine IoVT, utilizing an ISP operations within a differentiable CNN framework fusion architecture optimized for uneven illumination. This two-stage strategy combines global enhancement with detail optimization, effectively improving image quality, especially in poorly lit areas. A contrastive language-image pretraining (CLIP)-based multimodal iterative optimization allows for unsupervised training of the enhancement algorithm. By integrating traditional image signal processing (ISP) with convolutional neural networks (CNN), our approach reduces computational complexity while maintaining high performance, making it suitable for real-time deployment on edge devices. Experimental results demonstrate that our method effectively mitigates uneven brightness and enhances key image quality metrics, with preservation of original visual information (PSNR) improvements of 2.9%–4.9%, structural similarity (SSIM) by 4.3%–11.4%, and visual information fidelity (VIF) by 4.9%–17.8% compared to seven state-of-the-art algorithms. Simulated coal mine monitoring scenarios validate our method’s ability to balance performance and computational demands, facilitating real-time enhancement and supporting safer mining operations.

Wearable device-based Human Activity Recognition (HAR) has attracted considerable interest with the rapid development of the Internet of things (IoT), and Federated Learning (FL) has been widely adopted in this domain for its ability to collaboratively train models across decentralized devices while preserving privacy. However, its performance is hindered by data heterogeneity arising from variations in the placement of the wearable devices, user behaviors, and physiological characteristics. In this work, we present LDA-FedHAR, a federated HAR framework designed for wearable devices by capturing more common knowledge from aligned client HAR data. It performs Local HAR Data Alignment (LDA) on each client, which is an entirely on-device alignment method that operates independently on local HAR data. By computing the transformation matrix solely from local HAR data and applying it to the data itself, LDA projects heterogeneous client data into a unified space, thereby reducing inter-client discrepancies at the source. To further enhance efficiency and robustness, we propose two IMU-specific variants, LDA(S-IMU) and LDA(C-IMU), which explore intra-and inter-IMU correlations based on practical placements of wearable devices. Experiments are conducted on 4 public HAR datasets: HHAR, Shoaib2014, OPPORTUNITY++, and PAMAP2. The results show that LDA effectively reduces inter-client discrepancies, and LDA-FedHAR along with its variants consistently outperforms state-of-the-art FL methods. Moreover, the improvements achieved by integrating LDA into other FL methods highlight its applicability.

Perceiving scene depth and 3-D structure is one of the key tasks for Internet of Video Things (IoVT) devices to understand and interact with the environment. Self-supervised monocular depth estimation has demonstrated significant potential in leveraging large-scale unlabeled datasets to achieve competitive performance, thereby playing an increasingly important role in depth estimation. Despite recent methods providing additional supervisory signals through self-distillation strategies to improve depth estimation, an effective method for generating pseudo-depth labels suitable for addressing occlusion issues among elements far from the camera remains unexplored. To address this limitation, we propose a patch-based self-distillation learning framework to exploit the potential of self-supervised monocular depth estimation in recovering fine-grained scene depth. In the proposed framework, elements far from the camera within the input image are enlarged by enlarging and cropping operations in the patch-based self-distillation branch. Guided by photometric consistency, the model learns the detailed occlusion relationships among elements from the enlarged patches, producing patch depth maps with fine structures. In the main branch, which takes full-scale images as input, patch depth maps serve as pseudo-depth labels through self-distillation loss to provide additional supervisory signals for regions where photometric consistency fails to offer effective supervision. This forces the depth estimation network to recover fine structures of elements far from the camera in full-scale input images. Regarding the architecture of the depth estimation network, we introduce a bin-center prediction. In this prediction, a global aggregator based on self-attention provides additional scene structure queries for adaptive scene depth discretization. Finally, to encourage the model to explore more general cues for depth inference beyond road plane cues, we propose a PatchMix data augmentation method to enhance the model’s generalization ability to unseen scenes. Extensive experiments on the KITTI dataset show that the proposed method significantly improves performance over the baseline, particularly in fine-grained scene depth estimation. Moreover, the model also exhibits good generalization performance when transferred to the Make3D and Cityscapes datasets.

For visual disbalance defects (VDDs) in low-light images, such as brightness unevenness and color imbalance, existing enhancement methods struggle to extract defect features from local regions and apply adaptive enhancement based on varying degrees of these defects. To address these challenges, we propose an unsupervised multi-modal enhancement method based on a high-order adaptive curve, named CLIP-AE. Specifically, we introduce a multi-modal recurrent optimization approach utilizing contrastive language-image pre-training (CLIP). This method iteratively optimizes variable embedded prompts and an Adaptive Enhancement Module (AEM) to establish dependencies between the prompts and detailed style features in the images, guiding the AEM to perform adaptive image enhancement. Additionally, we implement a progressive feature alignment strategy to enhance the model's ability to perceive style features and improve optimization efficiency by using multiple enhanced images with identical content features and incremental style features. In the AEM, the optimized Hyperparameters Generative Network (HGN) generates the optimal hyperparameters, which drive a High-Dimensional Nested Gamma correction (HDN-Gamma) to perform pixel-wise adaptive enhancement for VDDs. HDN-Gamma further maps pixel values using specific enhancement curves to avoid artifacts. Extensive experiments demonstrate that our method effectively improves visual disbalance defects and reduces artifacts. Compared to seven state-of-the-art algorithms, our method shows significant improvements (PSNR: 16.46%, 16.89%, and 15.14%; SSIM: 9.26%, 8.02%, and 9.85%; MUSIQ: 6.37%, 6.54%, and 7.45%) on the LOL, SICE, and MIT-Adobe FiveK datasets. Our approach offers a novel solution for applying multimedia technology in low-light image enhancement tasks.

In the context of the digital age, utilizing cutting-edge technology for the digitization and creative expansion of ancient murals is crucial, aimed at preserving and passing on cultural heritage. Existing image outpainting techniques suffer from a lack of semantic guidance. This article introduces MuralAgent, a multimodal model based on Retrieval-Augmented Generation (RAG) technology. It precisely extracts key information from mural images and integrates it with a constructed ancient texts knowledge base to ensure the cultural and semantic consistency of the expanded images. Moreover, fine-tuning the Stable Diffusion model ensures the fidelity of the generated image styles. Specifically, this study involves constructing an ancient texts knowledge base for accurate matching, designing specific prompts for GPT-4V(ision) to extract key information, and innovatively expanding artworks through Stable Diffusion, providing a novel way for the public to reinterpret ancient murals.

Low-light image enhancement is a crucial visual task, and many unsupervised methods overlook the degradation of visible information in low-light scenes, adversely affecting the fusion of complementary information and hindering the generation of satisfactory results. To address this, we introduce Wakeup-Darkness, a multimodal enhancement framework that innovatively enriches user interaction through voice and textual commands. This approach signifies a technical leap and represents a paradigm shift in user engagement. We introduce a Cross-Modal Feature Fusion (CMFF) that synergizes semantic and depth context with low-light enhancement operations. Moreover, we propose a Gated Residual Block (GRB) and a channel-aware Look-Up Table (LUT) to adjust the intensity distribution of each channel. Crucially, the proposed Wakeup-Darkness scheme demonstrates remarkable generalization in unsupervised scenarios. The source code can be accessed from https://github.com/zhangbaijin/Wakeup-Dakness .

BACKGROUND: As a key non-coding RNA molecule, miRNA profoundly affects gene expression regulation and connects to the pathological processes of several kinds of human diseases. However, conventional experimental methods for validating miRNA-disease associations are laborious. Consequently, the development of efficient and reliable computational prediction models is crucial for the identification and validation of these associations. RESULTS: In this research, we developed the PCACFMDA method to predict the potential associations between miRNAs and diseases. To construct a multidimensional feature matrix, we consider the fusion similarities of miRNA and disease and miRNA-disease pairs. We then use principal component analysis(PCA) to reduce data complexity and extract low-dimensional features. Subsequently, a tuned cascade forest is used to mine the features and output prediction scores deeply. The results of the 5-fold cross-validation using the HMDD v2.0 database indicate that the PCACFMDA algorithm achieved an AUC of 98.56%. Additionally, we perform case studies on breast, esophageal and lung neoplasms. The findings revealed that the top 50 miRNAs most strongly linked to each disease have been validated. CONCLUSIONS: Based on PCA and optimized cascade forests, we propose the PCACFMDA model for predicting undiscovered miRNA-disease associations. The experimental results demonstrate superior prediction performance and commendable stability. Consequently, the PCACFMDA is a potent instrument for in-depth exploration of miRNA-disease associations.

Abstract Existing image enhancement algorithms often fail to effectively address issues of visual disbalance, such as brightness unevenness and color distortion, in low‐light images. To overcome these challenges, we propose a TransISP‐based image enhancement method specifically designed for low‐light images. To mitigate color distortion, we design dual encoders based on decoupled representation learning, which enable complete decoupling of the reflection and illumination components, thereby preventing mutual interference during the image enhancement process. To address brightness unevenness, we introduce CNNformer, a hybrid model combining CNN and Transformer. This model efficiently captures local details and long‐distance dependencies between pixels, contributing to the enhancement of brightness features across various local regions. Additionally, we integrate traditional image signal processing algorithms to achieve efficient color correction and denoising of the reflection component. Furthermore, we employ a generative adversarial network (GAN) as the overarching framework to facilitate unsupervised learning. The experimental results show that, compared with six SOTA image enhancement algorithms, our method obtains significant improvement in evaluation indexes (e.g., on LOL, PSNR: 15.59%, SSIM: 9.77%, VIF: 9.65%), and it can improve visual disbalance defects in low‐light images captured from real‐world coal mine underground scenarios.

With the rapid development of Internet of Things (IoT) and UAV technology, for the whole IoT system of vehicle detection, the middle and high level of information transmission and server processing has made a breakthrough, so at the bottom of the real-time detection of the vehicle by the UAV is the key to the whole system. However, UAV vehicle detection faces the challenges of too many small targets in the image leading to low detection accuracy, limited hardware platform resources requiring control of model size, and real-time detection requiring high inference speed. Aiming at the above problems, we propose a lightweight model PrFu-YOLO based on YOLOv8 improvement, which achieves a good balance between the accuracy, inference speed, and model size. And it realizes real-time vehicle detection embedded in an UAV platform. To solve the problem of low vehicle detection accuracy, we design a new structure PrFuFPN based on adding a small target detection layer to achieve more advanced feature fusion. To address the limited resources of the platform and the problem of real-time vehicle detection, we add GhostConv to the structure and constantly try to adjust the parameters of the network. Finally, extensive experiments were conducted on the VisDrone2019 and CARPK data sets to fully evaluate the model. Compared to YOLOv8s on the VisDrone2019 test set, mAP50 was improved by 10.05%, mA95 by 14.54%, the number of parameters was reduced by 8.13%, and the model size was reduced by 12.5%, while the FPS of 67.13 fully met the needs of real-time detection.

The power inspection is an important application of UAV-assisted power internet of video things (IoVT) for maintaining the safety of the power system. Due to the limitations of distance and angle, the resolution of the images captured by UAV is low, which seriously impacts the effects of small insulator defects detection. To address this problem, we propose a small-size defects detection method based on multi-scale feature interaction transformer for UAV-assisted Power IoVT. For the algorithm, we design a super-resolution reconstruction-assisted small object detection algorithm, the super-resolution module generates high-resolution images with the requirements of object detection function, which greatly improves the small object detection performance. Moreover, we design multi-scale feature interaction transformer network (MFITN), compared with the traditional non-local attention mechanism, the network structure can capture dependencies in multi-scales features, furthermore, the advantage assist the super-resolution module to generate more realistic image information to further improve small object detection. In addition, we propose a distributed model deployment strategy to deploy our high computational complexity algorithm in the edge side of the IoVT system, which can drive the overall algorithm to perform low-latency edge computation by relying only on the limited computing power devices. Experiments demonstrate that our method has better small object detection performance (mAP=81.3%, FPS=49.7), the super-resolution reconstruction is able to recover more realistic detail information, the distributed computing method can reduce the response latency by 33.4%-87.2%, which all contribute UAV-assisted Power IoVT system to realize accurate and fast power insulator defects detection.

Video surveillance as an important function of internet of video things (IoVT) system has been widely used in coal mine monitoring for coal mine safety with excellent results, however, there are still many shortcomings: 1) Existing coal mine IoVT systems have limited detection accuracy for small-sized objects; 2) Coal mine video surveillance systems generally adopt centralized cloud computing, transmission of massive data causes high latency, which seriously affects the response speed of object detection function; 3) The concept drift caused by the data stream seriously affect the detection effect of the offline algorithm. To address the above issues, we propose a small object detection method based federated intelligence to assist coal mine IoVT for object detection. First, we design a lightweight neural network Rep-ShuffleNet to improve YOLOv8, the state-of-the-art YOLO algorithm, to maintain high detection accuracy while dramatically increasing the inference speed, and with the advantage of lightweight, it can be deployed to embedded devices for low-latency edge computing; Moreover, we design a federated learning-based MLC-FL algorithm for local algorithms’ automatic and efficient optimization by asynchronous communication and data interaction reduction strategy. The experimental results show that with the assistance of federated intelligence model optimization strategies, the lightweight YOLOv8 has excellent detection performance (mAP: 94.6%, APsmall: 86.7%, FPS: 21.6), thus to assist coal mine IoVT to realize accurate and real-time underground small object detection.

The growing utilization of unmanned aerial vehicles (UAVs) across diverse industries has led to increased interest in UAV-assisted data acquisition for the Internet of Things (IoT). The security of image data collected by UAVs during transmission within the IoT has become a critical concern. This article focuses on the security challenges associated with UAV-assisted image data collection in the IoT and presents a dedicated framework designed to enhance the security of this process. Given the high-resolution nature of UAV-captured images, traditional encryption methods face difficulties in directly and effectively encrypting such data. To address this issue, this article introduces an efficient chaotic image encryption algorithm integrated into the proposed protection framework. The algorithm features a novel 1-D chaotic system for generating effective chaotic sequences. For the scrambling phase, a chaotic four-spiral transformation method is employed, and the diffusion process utilizes the Fibonacci matrix. This strategic approach aims to minimize pixel correlation within the image, thereby bolstering the overall security of the encryption process. Experimental validation conducted on authentic UAV image data sets demonstrates the superior, practical, secure, and efficient characteristics of the proposed algorithm.

The authenticity and integrity of sensed data in the data collection stage is a very critical aspect for the smart city industrial environment. They impact the accuracy of data analysis and the objectivity of making decisions. However, how to identify attack behaviors from environmental interference and establish a secure route to transmit data for resource-constrained terminals are challenging problems. To address these problems, in this article, we propose a trust-based security system (TSS). In TSS, we first design a trust model using binomial distribution for calculating the node's trust value and a third-party recommendation scheme for improving the objectivity of trust value. Then, we propose a trust management scheme for preventing theon–offattack. After that, we design a secure routing protocol, which is used to balance the security, transmission performance, and energy efficiency. Finally, the analytical results of the TSS are evaluated with extensive simulation experiment.

The rapid development of the Internet of Things (IoT) has brought a data explosion and a new set of challenges. It has been an emergency to construct a more robust and precise model to predict the electricity consumption data collected from the Internet of Things (IoT). Accurately forecasting the electricity consumption is a crucial technology for the planning of the energy resource which could lead to remarkable conservation of the building electricity consumption. This paper is focused on the electricity consumption forecasting of an office building with a small‐scale dataset, and 117 daily electricity consumption of the building are involved in the dataset, among which 89 values are selected as the training dataset and the remaining 28 values as the testing dataset. The hybrid model ARIMA (autoregression integrated moving average)‐SVR (support vector regression) is proposed to predict the electricity consumption with different prediction horizons ranging from 1 day to 28 days. The model performances are assessed by three evaluation indicators, respectively, are the mean squared error (MSE), the root mean square error (RMSE), and the mean absolute percentage error (MAPE). The proposed model ARIMA‐SVR is compared with the other four models, respectively, are the ARIMA, ARIMA‐GBR (gradient boosting regression), LSTM (long short‐term memory), and GRU (gated recurrent unit) models. The experiment result shows that the ARIMA‐SVR model has lower prediction errors when the prediction horizon is within 20 days, and the ARIMA model is better when the prediction horizon is in the interval of 20 to 28 days. The provided method ARIMA‐SVR has higher flexibility, and it is a great choice for electricity consumption prediction with more accurate results.

Wearable devices provide an extremely convenient way to collect a large amount of human motion data. In this paper, the human motion recognition method based on wearable devices is studied. We smooth the data to remove the noise caused by additional motion first. After that, the characteristic values that can distinguish the types of activities can be extracted. Then, we propose a human motion recognition method based on the improved one-dimensional convolutional neural networks(1D-CNNs). Compared with other traditional classification and recognition methods, the recognition rates of 11 human motions have been greatly improved. The average accuracy of each activity identification can reach 92.8%, while the average precision and recall can reach 98.7% and 92.8%.

Schizophrenia (SZ) is a complex human disease. It is a neurodegenerative disease characterized by the gradual loss of brain function, especially memory and cognitive ability. For many years, MRI has been widely used in schizophrenia studies because it can recognize structural and functional abnormalities in the brain region. In recent years, the most important research topic in the study of mental illness is to predict the cognitive performance of subjects from magnetic resonance imaging (MRI) measurements, and also include the recognition of related imaging biomarkers. Traditionally, this task has been a linear regression problem, but most existing studies cannot capture the relation-ship between the complex nonlinear cognitive properties and MRI measures. Inspired by these observations, we propose a Nonlinear Sparse Bayesian Learning (NSBL) model, and construct a sparse multivariate algorithm. Unlike the existing sparse algorithm, in our proposed model, the nonlinear function of the prediction matrix is responded by extending the block structure. The results show that the nonlinear sparse regression model can obtain better prediction accuracy. The model can use the correlation coefficient vector between vectors, it can also use the intra-block correlation in each regression coefficient.

Objective To explore the correlation between imaging data and genetic data of schizophrenia using group sparse canonical correlation analysis method. Methods A group sparse canonical analysis method was proposed, group sparse constraints λ1∥U∥Gand λ2∥U∥Gwere added to traditional canonical correlation analysis model to select features groups. Then, features within each group were selected by sparse constraints τ1∥U∥1and τ2∥V∥1. The group sparse canonical correlation analysis method was used to analyze the correlation between brain regions and genes of schizophrenia, and it's stability and ability were also verified to select biomarkers. Results Several pairs canonical brain regions and genes were identified. The left insula and gene AKT1 produced the most significant correlation, r=0.6538, and the correlations between both right rectus and gene DAOA, MAGI2 were more than 0.6. The correlation coefficients of selected features were (0.626 9±0.016 1) with group sparse canonical correlation analysis and (0.625 5±0.018 1) with sparse canonical correlation analysis. The biomarkers selected by group sparse canonical correlation analysis using 75 related genes to schizophrenia, was higher than using non-related genes randomly. Conclusion Several pairs canonical brain regions and genes can be identified by the group sparse canonical analysis method, which provided a new way for the study of schizophrenia and other complex mental disorders.

We in this paper consider a scenario where multiple vehicles jointly provision computing resources to obtain their benefits in the contexts of vehicular fog computing. A community that vehicles can freely join and leave is sponsored by a road side unit (RSU) and thus a resource pool is established such that tasks can be performed by sufficient computing resources. RSU as a coordinator takes in charge of decision making for task scheduling. A permutation of community members is established in advance and updated periodically so as to make the most suitable decision. A task scheduling strategy is proposed from the perspective of service oriented architecture. We have carried out the experiments to investigate our approach and the experimental results have revealed our approach has a great advantage over other approaches in terms of pursuing the values of the community.

As a key component of the information sensing and aggregating for big data, cloud computing, and Internet of Things (IoT), the information security in wireless sensor network (WSN) is critical. Due to constrained resources of sensor node, WSN is becoming a vulnerable target to many security attacks. Compared to external attacks, it is more difficult to defend against internal attacks. The former can be defended by using encryption and authentication schemes. However, this is invalid for the latter, which can obtain all keys of the network. The studies have proved that the trust management technology is one of effective approaches for detecting and defending against internal attacks. Hence, it is necessary to investigate and review the attack and defense with trust management. In this paper, the state-of-the-art trust management schemes are deeply investigated for WSN. Moreover, their advantages and disadvantages are symmetrically compared and analyzed in defending against internal attacks. The future directions of trust management are further provided. Finally, the conclusions and prospects are given.

With the rapid development of the Internet of things (IoT) technology, the application of IoT has been expanded greatly, and the disadvantages of the traditional battery power supply have become increasingly prominent. The power supply mode limits the development of the concrete structural health monitoring network. And the application of magnetic resonance coupled wireless power transfer technology can solve the problem of power supply to sensors embedded in concrete. The corrected transmission efficiency considering the concrete conductivity is proposed which establishes the relationship between the electromagnetic field and the circuit model. And the field-circuit coupled model of asymmetric wireless power transfer system in concrete is developed. The effects of radial offset and axial dislocation on the transmission efficiency at different concrete conductivity are further analyzed. The relationship between the resonant frequency and the transmission efficiency in different concrete conductivity is analyzed, and an optimization scheme is proposed to improve the transmission efficiency. Finally, the experimental setups are established, and the theoretical analysis is verified. The conclusions cannot only break through the bottleneck of the scale of the concrete structural health monitoring network but also further releases the application potential of IoT.

In this paper, we propose a cooperative model combined the multi-task reverse sparse representation model (MTRSR) and the AdaBoost classifier, which were used to cope with the disturbing of target gradient information caused by motion blur or target serious occlusion, and a descriptive dictionary were used to estimate the weights of each candidates. First, we use the MTRSR model to get the blur kernel which were used to get the blur target template set, meanwhile the confidence of the candidates is also obtained by the reconstruction error. Then we use the HOG features of the target templates to get the descriptive dictionary to calculate the weights of the candidates, and a AdaBoost classifier is used to calculate the confidences of all candidates. Finally, the best target is retrieved by the sum of production of weight value and the two confidences. The experimental data show that the proposed algorithm can fully cope with the target's information change which were caused by motion blur and target occlusion in the complex scene, and our algorithm can further improve the accuracy and robustness in visual tracking.

Healthcare-oriented wireless sensor network (HWSN) is one of the applications of wireless sensor networks in e-health. It not only can better achieve the physiological information of people, but also more efficiently reduce the Iatency regarding information collection and transmission. However, similar to other distributed networks, it also faces enormous security challenges, especially from internal attacks. It is difficult to distinguish many attack behaviors from interference in the complex healthcare scenarios, such as On-Off attack. In this paper, we propose a Binomial Distribution-based Trust Management Scheme (BDTMS) for HWSN. The proposed method can rapidly detect and effectively defend against On-Off attacks. In addition, the proposed method is also applicable to defending against bad mouthing attacks. Simulation results show that, compared with the Time-window-based Resilient Trust Management Scheme (TRTMS), our proposed BDTMS achieves better performance in defending against On-Off attack under obstacle movement, especially with higher detection accuracy.

Wireless sensor networks (WSNs) receive a lot of attention because of their great potential in monitoring the physical or environmental conditions of military, industry, and civilian. Moreover, mobile cloud computing (MCC) is widely focused, as they can greatly alleviate the hardware limit of mobile devices as well as enable a lot of new mobile applications. All these make the integration of WSNs and MCC a very hot research topic. In this paper, we first observe a context non-awareness issue between mobile user and WSNs, which affects the mobile user obtaining the desirable data when integrating WSNs and MCC. Then focusing on solving the context non-awareness issue to provide desirable data to mobile users, we propose a novel framework for integrating WSNs and MCC. The proposed framework performs data recommendation, data prediction as well as data traffic monitoring in the cloud to obtain the data feature information required by the mobile users and potential status of WSNs. Then these user data feature information and potential WSNs status information are utilized to optimize the deployment of WSNs and check the status of WSNs. This could in turn offer the desirable data to the mobile users. Extensive evaluations also validate the effectiveness of the proposed framework.