检索结果-内蒙古大学图书馆

Domain generalization with semi-supervised learning for people-centric activity recognition

Science China(information Sciences) 2025年第1期68卷 171-188页

作者： Jing LIU Wei ZHU Di LI Xing HU Liang SONG Academy for Engineering & Technology Fudan University Shanghai East-bund Research Institute on Networking Systems of AI School of Optoelectronic Information and Computer Engineering University of Shanghai for Science & Technology

People-centric activity recognition is one of the most critical technologies in a wide range of real-world applications,including intelligent transportation systems, healthcare services, and brain-computer interfaces. Large-scale data collection and annotation make the application of machine learning algorithms prohibitively expensive when adapting to new tasks. One way of circumventing this limitation is to train the model in a semi-supervised learning manner that utilizes a percentage of unlabeled data to reduce the labeling burden in prediction tasks. Despite their appeal, these models often assume that labeled and unlabeled data come from similar distributions, which leads to the domain shift problem caused by the presence of distribution gaps. To address these limitations, we propose herein a novel method for people-centric activity recognition,called domain generalization with semi-supervised learning(DGSSL), that effectively enhances the representation learning and domain alignment capabilities of a model. We first design a new autoregressive discriminator for adversarial training between unlabeled and labeled source domains, extracting domain-specific features to reduce the distribution gaps. Second, we introduce two reconstruction tasks to capture the task-specific features to avoid losing information related to representation learning while maintaining task-specific consistency. Finally, benefiting from the collaborative optimization of these two tasks, the model can accurately predict both the domain and category labels of the source domains for the classification task. We conduct extensive experiments on three real-world sensing datasets. The experimental results show that DGSSL surpasses the three state-of-the-art methods with better performance and generalization.

关键词： activity recognition deep learning domain generalization semi-supervised learning adversarial training

来源：评论

学校读者我要写书评

暂无评论

Multi-radio fast routing recovery protocol based on spectrum situation awareness

引用

Chinese Journal of Aeronautics 2025年第3期38卷 435-446页

作者： Chuanxin ZHOU Chao DONG Xiaojun ZHU Fuhui ZHOU Qihui WU Zhengrui QIN Ministry of Industry and Information Technology Nanjing University of Aeronautics and AstronauticsNanjing 211106China College of Computer Science and Technology Nanjing University of Aeronautics and AstronauticsNanjing 211106China School of Computer Science and Information Systems Northwest Missouri State UniversityMaryville 64468USA

UAV networks often encounter jamming attacks, under which multi-radio protocols have to switch radios to accelerate communication recovery. However, the existing protocols rely on exchange of hello messages to detect jamming, leading to long sensing time and thus slow routing recovery. To address the issues raised by jamming attacks, we propose a new routing protocol, Electromagnetic Spectrum situation awareness Optimized Link State Routing (ESOLSR) protocol, to improve the existing OLSRv2 protocol. ESOLSR utilizes the spectrum situation awareness capability from the physical layer, and adopts joint-updating of link status, updating of interface functions, and adaptive adjustment of parameters. Our simulation results show that the improved protocol, ESOLSR, can recover routing and resume normal communication 26.6% faster compared to the existing protocols.

关键词： UAV ad hoc network OLSR protocol Routing recovery Electromagnetic spectrum situation awareness Multi-radio

来源：评论

学校读者我要写书评

暂无评论

Classifying distinct emotions from parents of ASD child using EEG source data by combining Bernoulli–Laplace Prior and graph neural networks

引用

Neural Computing and Applications 2025年第12期37卷 7877-7895页

作者： ArulDass, Stephen Dass Jayagopal, Prabhu School of Computer Science Engineering and Information Systems Vellore Institute of Technology Tamil Nadu Vellore632014 India

Emotion recognition using biological brain signals needs to be reliable to attain effective signal processing and feature extraction techniques. The impact of emotions in interpretations, conversations, and decision-making, has made automatic emotion recognition and examination of a significant feature in the field of psychiatric disease treatment and cure. The problem arises from the limited spatial resolution of EEG recorders. Predetermined quantities of electroencephalography (EEG) channels are used by existing algorithms, which combine several methods to extract significant data. The major intention of this study was to focus on enhancing the efficiency of recognizing emotions using signals from the brain through an experimental, adaptive selective channel selection approach that recognizes that brain function shows distinctive behaviors that vary from one individual to another individual and from one state of emotions to another. We apply a Bernoulli–Laplace-based Bayesian model to map each emotion from the scalp senses to brain sources to resolve this issue of emotion mapping. The standard low-resolution electromagnetic tomography (sLORETA) technique is employed to instantiate the source signals. We employed a progressive graph convolutional neural network (PG-CNN) to identify the sources of the suggested localization model and the emotional EEG as the main graph nodes. In this study, the proposed framework uses a PG-CNN adjacency matrix to express the connectivity between the EEG source signals and the matrix. Research on an EEG dataset of parents of an ASD (autism spectrum disorder) child has been utilized to investigate the ways of parenting of the child's mother and father. We engage with identifying the personality of parental behaviors when regulating the child and supervising his or her daily activities. These recorded datasets incorporated by the proposed method identify five emotions from brain source modeling, which significantly improves the accurac

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Employing cross-domain modelings for robust object detection in dynamic environment of autonomous vehicles

引用

Multimedia Tools and Applications 2024年 1-47页

作者： Rawlley, Oshin Gupta, Shashank Kathera, Hardik Katyal, Siddharth Batwara, Yashvardhan Department of Computer Science & amp Information Systems Birla Institute of Technology and Science Rajasthan Pilani India

Object detection (OD) in Advanced Driver Assistant systems (ADAS) has been a fundamental problem especially when complex unseen cross-domain adaptations occur in real driving scenarios of autonomous Vehicles (AVs). During the sensory perception of autonomous Vehicles (AV) in the driving environment, the Deep Neural Networks (DNNs) trained on the existing large datasets fail to detect the vehicular instances in the real-world driving scenes having sophisticated dynamics. Recent advances in Generative Adversarial Networks (GAN) have been effective in generating different domain adaptations under various operational conditions of AVs, however, it lacks key-object preservation during the image-to-image translation process. Moreover, high translation discrepancy has been observed with many existing GAN frameworks when encountered with large and complex domain shifts such as night, rain, fog, etc. resulting in an increased number of false positives during vehicle detection. Motivated by the above challenges, we propose COPGAN, a cycle-object preserving cross-domain GAN framework that generates diverse variations of cross-domain mappings by translating the driving conditions of AV to a desired target domain while preserving the key objects. We fine-tune the COPGAN training with an initial step of key-feature selection so that we realize the instance-aware image translation model. It introduces a cycle-consistency loss to produce instance specific translated images in various domains. As compared to the baseline models that needed a pixel-level identification for preserving the object features, COPGAN requires instance-level annotations that are easier to acquire. We test the robustness of the object detectors SSD, Detectron, and YOLOv5 (SDY) against the synthetically-generated COPGAN images, along with AdaIN images, stylized renderings, and augmented images. The robustness of COPGAN is measured in mean performance degradation for the distorted test set (at IoU threshold =

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Robust steganographic approach using generative adversarial network and compressive autoencoder

引用

Multimedia Tools and Applications 2024年 1-38页

作者： Qasaimeh, Malik Qtaish, Alaa Abu Aljawarneh, Shadi Department of Computer Information Systems Faculty of Computer and Information Technology Jordan University of Science and Technology Irbid Jordan Department of Software Engineering co-joint with the Department of Cyber Security Faculty of Computer and Information Technology Jordan University of Science and Technology Irbid Jordan

Nowadays, social media applications and websites have become a crucial part of people’s lives;for sharing their moments, contacting their families and friends, or even for their jobs. However, the fact that these valuable data are transferred via the Internet and open channels, which are vulnerable to attacks and espionage, requires using defense methods to improve the security of data transmission. Recently, the progress of Deep Learning (DL) inspired researchers to use it with security methods such as steganography, the art of hiding secret data in unrelated content. DL and steganography combined enhanced the hiding properties, especially in the Coverless Image Steganography (CIS) methods. In this paper, we propose a Compressive Coverless Image Steganography (CCIS) model, which is a generated-based CIS, to improve the capacity and robustness of Steganography. This model uses the compressive autoencoder to compress the message, thus increasing the capacity, and uses the Generative Adversarial Network (GAN) to generate a stego image from the compressed vector, thus enhancing concealment ability, besides using a regression model and optimization method to extract the data. Another version was designed to enable sending binary messages by replacing the compressive autoencoder network with a robust mapping rule. Experiments show that the proposed model could improve the characteristics of Steganography. Furthermore, the proposed model could extract binary messages with 256 bits from attacked stego images with 99% recovery accuracy. Thus, capacity and robustness were enhanced using our model with both images and text as secret messages. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Abnormal event detection in surveillance videos through LSTM auto-encoding and local minima assistance

引用

Discover Internet of Things 2025年第1期5卷 1-25页

作者： Sengonul, Erkan Samet, Refik Abu Al-Haija, Qasem Alqahtani, Ali Alsemmeari, Rayan A. Alghamdi, Bandar Alturki, Badraddin Alsulami, Abdulaziz A. Department of Computer Engineering Ankara University Ankara06100 Turkey Department of Cybersecurity Faculty of Computer & Information Technology Jordan University of Science and Technology PO Box 3030 Irbid22110 Jordan Department of Networks & Communications Engineering College of Computer Science & Information Systems Najran University Najran61441 Saudi Arabia Department of Information Technology Faculty of Computing and Information Technology King Abdulaziz University Jeddah21589 Saudi Arabia Department of Information Systems Faculty of Computing and Information Technology King Abdulaziz University Jeddah21589 Saudi Arabia

Abnormal event detection in video surveillance is critical for security, traffic management, and industrial monitoring applications. This paper introduces an innovative methodology for anomaly detection in video data, encompassing three primary stages: preprocessing, feature learning, and anomaly detection. We employ background subtraction and noise reduction during preprocessing to refine the data. The feature-learning stage involves training an LSTM autoencoder to capture the essential features of normal video sequences. For anomaly detection, we map video data to a lower-dimensional space (latent code) and compare it against the distribution of codes from normal sequences. We determine regularity scores and identify local minima points exceeding a specified threshold while scrutinizing shadows between adjacent maxima to confirm and pinpoint anomalies. When tested on the CUHK Avenue and UMN datasets, our methodology demonstrated performance with AUCs of 93.8% and 94.1%, respectively, outperforming several baseline models. Our results show the high precision of our method that can detect anomalies, highlighting its potential advantages that it can achieve for enhancing systems of surveillance. © The Author(s) 2025.

关键词： Noise abatement

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Approach to Industrial Corrosion Detection

引用

computers, Materials & Continua 2024年第11期81卷 2587-2605页

作者： Mehwash Farooqui Atta Rahman Latifa Alsuliman Zainab Alsaif Fatimah Albaik Cadi Alshammari Razan Sharaf Sunday Olatunji Sara Waslallah Althubaiti Hina Gull Department of Computer Engineering(CE) College of Computer Science and Information Technology(CCSIT)Imam Abdulrahman Bin Faisal UniversityDammam31441Saudi Arabia Department of Computer Science(CS) College of Computer Science and Information Technology(CCSIT)Imam Abdulrahman Bin Faisal UniversityDammam31441Saudi Arabia Department of Computer Information Systems(CIS) College of Computer Science and Information Technology(CCSIT)Imam Abdulrahman Bin Faisal UniversityDammam31441Saudi Arabia

The proposed study focuses on the critical issue of corrosion,which leads to significant economic losses and safety risks worldwide.A key area of emphasis is the accuracy of corrosion detection *** recent studies have made progress,a common challenge is the low accuracy of existing detection *** models often struggle to reliably identify corrosion tendencies,which are crucial for minimizing industrial risks and optimizing resource *** proposed study introduces an innovative approach that significantly improves the accuracy of corrosion detection using a convolutional neural network(CNN),as well as two pretrained models,namely YOLOv8 and *** leveraging advanced technologies and methodologies,we have achieved high accuracies in identifying and managing the hazards associated with corrosion across various industrial *** advancement not only supports the overarching goals of enhancing safety and efficiency,but also sets a new benchmark for future research in the *** results demonstrate a significant improvement in the ability to detect and mitigate corrosion-related concerns,providing a more accurate and comprehensive solution for industries facing these *** CNN and EfficientNetB0 exhibited 100%accuracy,precision,recall,and F1-score,followed by YOLOv8 with respective metrics of 95%,100%,90%,and 94.74%.Our approach outperformed state-of-the-art with similar datasets and methodologies.

关键词： Deep learning YOLOv8 EfficientNetB0 CNN corrosion detection Industry 4.0 sustainability

来源：评论

学校读者我要写书评

暂无评论

A Robust Tuned Random Forest Classifier Using Randomized Grid Search to Predict Coronary Artery Diseases

引用

computers, Materials & Continua 2023年第5期75卷 4633-4648页

作者： Sameh Abd El-Ghany A.A.Abd El-Aziz Department of Information Systems College of Computer and Information SciencesJouf UniversityKSA Department of Information Systems Faculty of Computers and InformationMansoura UniversityEgypt Department of Information Systems and Technology Faculty of Graduate Studies for Statistical ResearchCairo UniversityEgypt

Coronary artery disease(CAD)is one of themost authentic cardiovascular afflictions because it is an uncommonly overwhelming heart *** breakdown of coronary cardiovascular disease is one of the principal sources of death all over *** deterioration is a challenge,especially in youthful and rural countries where there is an absence of humantrained *** heart diseases happen without apparent signs,high-level detection is *** paper proposed a robust and tuned random forest model using the randomized grid search technique to *** proposed framework increases the ability of CADpredictions by tracking down risk pointers and learning the confusing joint efforts between ***,the healthcare industry has a lot of data but needs to gain more *** proposed framework is used for extracting knowledge from data stores and using that knowledge to help doctors accurately and effectively diagnose heart disease(HD).We evaluated the proposed framework over two public databases,Cleveland and Framingham *** datasets were preprocessed by using a cleaning technique,a normalization technique,and an outlier detection ***,the principal component analysis(PCA)algorithm was utilized to lessen the feature dimensionality of the two ***,we used a hyperparameter tuning technique,randomized grid search,to tune a random forest(RF)machine learning(ML)*** randomized grid search selected the best parameters and got the ideal CAD *** proposed framework was evaluated and compared with traditional *** proposed framework’s accuracy,sensitivity,precision,specificity,and f1-score were 100%.The evaluation of the proposed framework showed that it is an unrivaled perceptive outcome with tuning as opposed to other ongoing existing frameworks.

关键词： Coronary artery disease tuned random forest randomized grid search classifier

来源：评论

学校读者我要写书评

暂无评论

Modern Standard Arabic speech disorders corpus for digital speech processing applications

引用

International Journal of Speech technology 2024年第1期27卷 157-170页

作者： Alqudah, Assal A. M. Alshraideh, Mohammad A. M. Abushariah, Mohammad A. M. Sharieh, Ahmad A. S. Department of Computer Science King Abdullah II School of Information Technology The University of Jordan Amman Jordan Department of Computer Information Systems King Abdullah II School of Information Technology The University of Jordan Amman Jordan Department of Computer Science Faculty of Science and Information Technology Al-Zaytoonah University of Jordan Amman Jordan

Digital speech processing applications including automatic speech recognition (ASR), speaker recognition, speech translation, and others, essentially require large volumes of speech data for training and testing purposes. Although there are available speech corpora, speech data for speakers suffering speech disorders are hardly available for many languages including Arabic language. Consequently, developing digital speech processing applications that target the entire society becomes hard due to the unavailability of speech corpora that contain sufficient speakers’ variations including healthy and disordered speech. This research presents our work towards developing a Modern Standard Arabic (MSA) speech corpus for speakers suffering distortion and substitution articulation disorders. The speech corpus was recorded by 40 (20 male and 20 female) Jordanian speakers who suffer either distortion or/and substitution articulation disorders. This speech corpus can be used for various applications including ASR, speech and hearing, and others. Part of this speech corpus is used for developing and evaluating an ASR for MSA using the Carnegie Mellon University (CMU) Pocketsphinx tools based on Mel-Frequency Cepstral Coefficients (MFCC) and Hidden Markov Model (HMM) techniques. Furthermore, Linear Discriminant Analysis (LDA) and Maximum Likelihood Linear Transform (MLLT) optimization techniques were applied. Using three different testing data sets, this work obtained 98.38% and 1.76% average word recognition correctness rate (WRCR) and average Word Error Rate (WER), respectively, for speaker-dependent and text-independent. For speaker-independent and text-dependent, this work obtained 99.37% and 0.68% average WRCR and average WER, respectively, whereas for speaker-independent and text-independent this work obtained 96.53% and 4.00% average WRCR and average WER, respectively. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Natur

关键词： Hidden Markov models

来源：评论

学校读者我要写书评

暂无评论

Interaction-aware Trust Management Scheme for IoT systems With Machine Learning-Based Attack Detection

引用

IEEE Internet of Things Journal 2025年第11期12卷 17169-17182页

作者： Farhat, Ali Eldosouky, Abdelrahman Ibnkahla, Mohamed Matrawy, Ashraf Carleton University Department of Systems and Computer Engineering OttawaON Canada Carleton University School of Information Technology OttawaON Canada

The recent Internet of Things (IoT) adoption has revolutionized various applications while introducing significant security and privacy challenges. Traditional security solutions are unsuitable for IoT systems due to their dynamicity, heterogeneity, and resource constraints. Trust-based solutions are emerging as promising alternatives due to their ability to track the dynamic behavior in IoT systems. However, existing trust management schemes are implemented at the device level, raising several challenges, including device modification that compromises certification and scalability, increased network overhead, and higher device resource utilization. To address these challenges, this paper proposes a novel trust management scheme that shifts its implementation to a higher layer in the IoT system, specifically to the IoT access layer (e.g., gateway). The proposed scheme establishes trust based on typical device interactions with the gateway without requiring additional information from the device. It relies on objective attributes spanning communication, security, and advanced dimensions to compute the trust value of an IoT device. Additionally, an Artificial Neural Network (ANN) is integrated to determine if the device acts maliciously or behaves normally. Simulation results demonstrate a notable improvement in the detection rate, primarily due to incorporating the proposed ANN, compared to the threshold-based approaches in the literature. Overall, the improvements highlight the significant advantage of the proposed scheme's robustness. © 2025 IEEE.

关键词： Gateways (computer networks)

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：