An International Journal on Multi-Sensor, Multi-Source Information Fusion

The journal is intended to present within a single forum all of the developments in the field of multi-sensor, multi-source, multi-process information fusion and thereby promote the synergism among the many disciplines that are contributing to its growth. The journal is the premier vehicle for disseminating information on all aspects of research and development in the field of information fusion. Articles are expected to emphasize one or more of the three facets: architectures, algorithms, and applications. Papers dealing with fundamental theoretical analyses as well as those demonstrating their application to real-world problems will be welcome. The journal publishes original papers, letters to the Editors and from time to time invited review articles, in all areas related to the information fusion arena including, but not limited to, the following suggested topics:

• Data/Image, Feature, Decision, and Multilevel Fusion
• Multi-classifier/Decision Systems
• Multi-Look Temporal Fusion
• Multi-Sensor, Multi-Source Fusion System Architectures
• Distributed and Wireless Sensor Networks
• Higher Level Fusion Topics Including Situation Awareness And Management
• Multi-Sensor Management and Real-Time Applications
• Adaptive And Self-Improving Fusion System Architectures
• Active, Passive, And Mixed Sensor Suites
• Multi-Sensor And Distributed Sensor System Design
• Fusion Learning In Imperfect, Imprecise And Incomplete Environments
• Intelligent Techniques For Fusion Processing
• Fusion System Design And Algorithmic Issues
• Fusion System Computational Resources and Demands Optimization
• Special Purpose Hardware Dedicated To Fusion Applications
• Mining Remotely Sensed Multi-Spectral/Hyper-Spectral Image Data Bases
• Information Fusion Applications in Intrusion Detection, Network Security, Information Security and Assurance arena
• Applications such as Robotics, Space, Bio-medical, Transportation, Economics, and Financial Information Systems
• Real-World Issues such as Computational Demands, Real-Time Constraints in the context of Fusion systems.

Special Issue on Multi-Sensor Data Fusion for Smart Healthcare
截稿日期: 2024-12-15

Medical data exists in several formats, from structured data and medical reports to 1D signals, 2D images and 3D volumes or even higher dimensional data such as temporal 3D sequences. Healthcare experts can make an auscultation and produce a report in text format; an electrocardiogram can be made and printed in time series format, an x- ray can be performed and saved as an image; a volume can be provided through an angiography; temporal information can be given by echocardiograms, 4D information can be extracted through flow MRI. Another typical source of variability is the existence of data from different time points, such as pre and post treatment, for instance. This high and diverse amount of information needs to be organized and mined in an appropriate way so that meaningful information can be extracted. In recent times, multimodal medical data fusion and analysis can combine salient information into a single source to ensure better diagnostic accuracy and assessment, and ultimately personalized healthcare services delivery. It is established that multi-sensor data fusion provides a valuable solution for healthcare tele monitoring requirements. These healthcare applications demand the usage of wide range of heterogeneous sensors/devices. Further, the multi-sensor fusion approaches increase the reliability and robustness of the healthcare systems by reducing the threatens posed by various malfunctions in the sensors and environment itself such as the power and communication. In addition, global decisions can be taken based on the dependability between the data sources. Finally, exploiting and fusing generative AI techniques for medical data generation could be fundamental to create new opportunities and tackle complex issues in the healthcare area.

Motivated by these facts, this special issue targets the researchers from both academia and industry to explore and share new ideas, approaches, theories and practices with focus on multi-sensor data fusion for smart healthcare.Authors are invited to submit high quality proposal on topics including, but not limited to:

    Multi-sensor fusion for healthcare
    Combining multiple models for medical data
    Combining multiple sources in medical data
    Data Fusion in the Internet of medical things
    AI techniques/algorithms for multi-source medical data fusion
    Deep learning for multi-source medical data processing
    Feature fusion for medical data
    Hierarchical models for medical information fusion
    Fusion algorithm for Bio-Inspired data
    Tensor schemes and constraint algorithms for health and medical data fusion
    Heterogeneous information fusion for healthcare
    Medical data fusion trust, security and privacy in trust evaluation
    Fusing generative AI techniques for generating healthcare data

Guest editors:

Giancarlo Fortino, PhD
University of Calabria, Arcavacata di Rende, Italy. 
g.fortino@unical.it

David Camacho, PhD
Universidad Politécnica Madrid, Madrid, Spain. david.camacho@upm.es

Min Chen, PhD
South China University of Technology, Guangzhou, China.minchen@ieee.org

Amit Kumar Singh, PhD
National Institute of Technology Patna, Patna, India.amit.singh@nitp.ac.in

Xiaohui Tao, PhD 
University of Southern Queensland, Toowoomba, Australia. 
Xiaohui.Tao@unisq.edu.au

Manuscript submission information:

The journal's submission platform (Editorial Manager®) will be available for receiving submissions to this Special Issue from July 1st, 2024. Please refer to the Guide for Authors to prepare your manuscript and select the article type of “VSI: Multi-Sensor Fusion” when submitting your manuscript online. Both the Guide for Authors and the submission portal could be found on the Journal Homepage: Information Fusion | Journal | ScienceDirect.com by Elsevier.

Timeline:

Submission Open Date *01/07/2024
Final Manuscript Submission Deadline *15/12/2024
Editorial Acceptance Deadline *31/03/2025

Keywords:

Sensor Fusion, Healthcare, Generative AI, Medical data, Deep Learning

Special Issue on Underwater Optical Applications of Multi-source Information Fusion
截稿日期: 2024-12-15

The exploration and study of underwater environments crucially depend on acquiring high-quality imagery, challenged by factors like limited visibility and light scattering. Multi-source information fusion stands out as a key strategy, integrating multi-modal data and multi-model outputs to significantly enhance the analysis of underwater images. This approach leverages diverse sensor data—such as sonar, optical, and infrared—and employs advanced machine learning and artificial intelligence technologies. It aims to improve the accuracy of object detection, classification, and environmental understanding. We invite scholars to share their innovative research on multi-source information fusion for enhancing underwater imaging and surveillance capabilities. Submissions are encouraged to explore multi-modal processing, multi-model fusion techniques, and the application of large models in underwater environment analysis. This special issue aims to highlight technological advancements and expand applications in underwater imaging, addressing the complex challenges of accurately capturing and analyzing underwater scenes.

This special issue will focus on the following topics:

    Create an underwater multi-modal dataset (open source);
    Develop high-quality standard datasets for water-related optical images, underwater object detection, and underwater video;
    Implementing constrained underwater image fusion using deep learning or traditional techniques;
    Enhance the image quality of complex underwater environments (murky, low-light, blurry, etc.) through information fusion;
    Implement underwater object detection using acoustic and optical data;
    Enhance the quality of water-related optical images using deep learning-based multi-modal data fusion;
    Enhance water-related optical image information using optical imaging models；
    Utilize data fusion in AUVs for marine environmental monitoring, aquaculture, and underwater security;
    Employ advanced technology to decouple the interlinked degradation clues in underwater environments.

Guest editors:

Guang-Yong Chen, PhD
Fuzhou University, Fuzhou, China
gychen@fzu.edu.cn
(Image enhancement/restoration, computational intelligence, system identification)

Lin Wang, PhD
University of Jinan, Jinan, China
wangplanet@gmail.com
(Systems and cybernetics, computational intelligence, computer vision)

Long Chen, PhD
University of Macau, Macau, China
longchen@umac.mo
(Machine learning, computer vision, pattern recognition)

Vivone Gemine, PhD
National Research Council, Tito, Italy
gemine.vivone@imaa.cnr.it
(Image fusion, statistical signal processing, deep learning)

C. L. Philip Chen, PhD
South China University of Technology, Guangzhou, China
philip.chen@ieee.org
(Machine learning, system identification, systems and cybernetics)

Manuscript submission information:

The journal's submission platform (Editorial Manager®) will be available for receiving submissions to this Special Issue from June 15th, 2024. Please refer to the Guide for Authors to prepare your manuscript and select the article type of “VSI: Underwater Optical APP” when submitting your manuscript online. Both the Guide for Authors and the submission portal could be found on the Journal Homepage: Information Fusion | Journal | ScienceDirect.com by Elsevier.

Timeline:

Submission Open Date *15/06/2024
Final Manuscript Submission Deadline *15/12/2024
Editorial Acceptance Deadline *15/04/2025

Keywords:

Underwater, Information Fusion, Multi-source, Multi-model, Computer vision

Special Issue on Data Fusion Approaches in Data-Centric AI for Developing Trustworthy AI Systems
截稿日期: 2024-12-15

Data-Centric AI, the process of continuously and systematically transforming imperfect data into high-quality data, is increasingly becoming the central piece of Trustworthy AI solutions. Within this new paradigm, data is treated dynamically through the AI development lifecycle (from data collection and exploration, to model training/testing, deployment, and monitoring), requiring the development of innovative strategies regarding scalable, continuous, and iterative Data Fusion. Exploring multiple sources of data can provide not only useful information for model training and prediction, but also important insights on other aspects related to the data and the intended system itself, such as privacy, fairness, and/or explainability constraints.

Although Data Fusion approaches have the potential to provide a trustworthy, unified, and enhanced view of the data, they add several layers of complexity to the data-centric paradigm, namely in what concerns systematically managing data quality, heterogeneity, consistency, and scalability throughout the entire AI lifecycle, which turns the development of Trustworthy AI approaches even more challenging.

This special issue explores the shift towards prioritising data over models, fostering the integration and development of data fusion approaches and techniques that enhance the quality, fairness, interpretability, robustness, and trustworthiness of AI systems across various domains and applications. We encourage researchers and practitioners to conceptualise solutions that revolve around improving data, not models, to ensure efficient and ethical downstream applications. We welcome both theoretical and practical contributions, as well as real-world applications, including healthcare, bioinformatics, military, smart cities, finance, education, retail, transportation, energy, climate, sustainability, agriculture, and sports.

The topics of interest include, but are not limited to:

    Data integration and distributed metadata retrieval
    Distributed counterfactual design
    Decision fusion for enhanced explainability
    Data quality assessment and accountability
    Data governance and regulatory compliance
    Secure meta-learning and transfer learning
    Transparent multimodal data fusion
    Feature fusion for missing data imputation
    Data fusion techniques for bias mitigation
    Scalable data fusion and processing workflows
    Privacy-preserving data fusion techniques
    Explainable data augmentation and outlier detection
    Automatic data preprocessing recommendation
    Fusion methods for privacy-preserving synthetic data
    Visualisation strategies for data fairness monitoring
    Distributed missing data imputation
    Human-in-the-loop data engineering
    Temporal fusion for time series synthetic data generation
    Data engineering prompting
    Fusion learning with imperfect data
    Distributed data auditing and evaluation practices
    Distributed labelling and benchmarking techniques
    Datasets for Data-Centric and Data Fusion development

Guest editors:

Miriam Seoane Santos, PhD
University of Porto and LIAAD, Porto, Portugal
(Email: miriam.santos@fc.up.pt)

Nathalie Japkowicz, PhD
University of American, Washington, District of Columbia, United States of America
(Email: japkowic@american.edu)

Ana Carolina Lorena, PhD
Aeronautic Institute of Technology, SÃO JOSÉ DOS CAMPOS, Brazil
(Email: aclorena@ita.br)

Pedro Henriques Abreu, PhD
University of Coimbra and CISUC, Coimbra, Portugal
(Email: pha@dei.uc.pt)

Manuscript submission information:

The journal's submission platform (Editorial Manager®) will be available for receiving submissions to this Special Issue from September 15th, 2024. Please refer to the Guide for Authors to prepare your manuscript and select the article type of “VSI: Data-Centric AI” when submitting your manuscript online. Both the Guide for Authors and the submission portal could be found on the Journal Homepage: Information Fusion | Journal | ScienceDirect.com by Elsevier.

Timeline:

Submission Open Date *15/09/2024
Final Manuscript Submission Deadline *15/12/2024
Editorial Acceptance Deadline *15/04/2025

Keywords:

data-centric ai; (trustworthy ai) OR (responsible ai); data quality; explainability; (data fairness) OR (bias mitigation); privacy-preserving techniques; synthetic data; (multimodal data fusion) OR (fusion learning); meta-learning; data complexity

Special Issue on Explainable AI in Industry 4.0 and 5.0
截稿日期: 2024-12-30

This Special Issue aims to gather papers that focus on integrating and applying eXplainable AI (xAI) in the context of Industry 4.0 and 5.0. The themes that will be explored include the use of XAI in modern industrial settings and the collaboration between humans and machines in Industry 5.0, with a particular focus on trustworthiness, transparency, and accountability. As AI becomes more prevalent in industries, it is crucial to understand its nuances, particularly in terms of explainability. While there have been publications on the revolutionary nature of Industry 4.0 and 5.0, there has been little exploration of the critical role of XAI in these sectors. Previous special issues have covered AI and industrial revolutions, but a concerted effort to comprehend the importance of XAI in these settings is lacking. This Special Issue is intended for both academic researchers and industrial practitioners, with the potential to promote innovative collaborations.

Topics of interest include (but are not limited to):

    Data and information fusion in the industrial XAI context
    Explainable systems fusing various sources of industrial information
    Exploring XAI in the performance and efficiency of industrial systems
    XAI for predictive maintenance
    Forecasting of product and process quality
    Explainable anomaly detection
    Root Cause Analysis, Causal Reasoning
    Automatic process optimization
    Industrial process monitoring and modelling
    Visual analytics and interactive machine learning
    Remaining Useful Life
    Decision-making assistance and resource optimization
    Planning under uncertainty
    Digital Twins for Predictive Maintenance
    Analysis of usage patterns
    AI transparency and accountability in smart factories
    Ethical considerations in industrial deployment of AI
    Industrial use cases for XAI (e.g., manufacturing, energy, transport)
    Challenges and future directions for XAI in the industry

Guest editors:

Rita P. Ribeiro, PhD
University Porto and INESC TEC, Porto, Portugal

João Gama, PhD
University Porto and INESC TEC, Porto, Portugal

Slawomir Nowaczyk, PhD
Halmstad University, Halmstad, Sweden 

Sepideh Pashami, PhD
Halmstad University, Halmstad, Sweden 

Manuscript submission information:

The journal's submission platform (Editorial Manager®) will be available for receiving submissions to this Special Issue from January 29th, 2024. Please refer to the Guide for Authors to prepare your manuscript and select the article type of “VSI: XAI in Industry 4.0 and 5.0” when submitting your manuscript online. Both the Guide for Authors and the submission portal could be found on the Journal Homepage: Information Fusion | Journal | ScienceDirect.com by Elsevier.

Timeline:

Submission Open Date *29/01/2024
Final Manuscript Submission Deadline *30/12/2024
Editorial Acceptance Deadline *31/04/2025

Keywords:

Predictive Maintenance; Industry 4.0; Explainable AI; Machine Learning

Special Issue on Mixture of Experts (MoE) and Ensemble Learning for Big Data
截稿日期: 2024-12-31

In the age of artificial intelligence and the digital economy, data analysis and model building are considered compulsory processes in many applications. In this process, machine learning algorithms become important and valuable tools because pattern recognition, knowledge discovery from data for decision-making, and models learned from data for intelligent operations are enabled by efficient and effective algorithms. As the complexity of the operational processes increases, the complexity and diversity of the collected data also increase, which results in more algorithms being developed to satisfy the needs of data analysis. There are several theoretical and technical challenges in developing efficient, scalable, and approximate computing methods for complex big data.

This special issue aims to address the most recent developments and research findings related to the intersection of AI, NLP, and big data computing, focusing on areas like mixture of experts (MoE), parallel and distributed processing, approximate computing, intelligent data analysis, dynamic information systems, ensemble learning, large language model (LLM), etc. This issue strives to provide a platform for researchers and practitioners worldwide, encouraging innovative, cutting-edge, and state-of-the-art theoretical and applied solutions that can tackle the challenges associated with big data computing. It endeavors to promote discussions, stimulate ideas, and foster collaborations towards developing advanced, scalable, and intelligent computing systems.

The articles will provide cross-disciplinary innovative research ideas and application results for mixture of experts (MoE) and ensemble learning for big data including novel theory, algorithms, and applications. Manuscript submissions are encouraged for a broad theme of related topics, including but not limited to:

    Emerging mixture of experts (MoE) model
    Ensemble distributed machine learning algorithms
    Simplification and compression techniques for machine learning models
    Data and model parallelism for large language model (LLM)
    Distributed computing for low-consumption computing
    Scalable data mining for big data analytics
    Energy-efficient AI models and applications
    Large and multi-modal data processing
    Feature fusion-based data augmentation in a resource-limited computing scenario
    Data-centric perspective to build efficient and trustworthy information service applications

Guest editors:

Mohammad Sultan Mahmud, Ph.D
Shenzhen University, Shenzhen, China. Email: sultan@szu.edu.cn

Diego Garcia-Gil, Ph.D
University of Granada, Granada, Spain. Email: djgarcia@decsai.ugr.es

Joshua Zhexue Huang, Ph.D
Shenzhen University, Shenzhen, China. Email: zx.huang@szu.edu.cn

Alladoumbaye Ngueilbaye, Ph.D
Shenzhen University, Shenzhen, China. Email: angueilbaye@szu.edu.cn

Xudong Sun, Ph.D
Shenzhen University, Shenzhen, China. Email: sunxudong2016@email.szu.edu.cn

Manuscript submission information:

The journal's submission platform (Editorial Manager®) will be available for receiving submissions to this Special Issue from July 8th, 2024. Please refer to the Guide for Authors to prepare your manuscript and select the article type of “VSI: MoE and Ensemble learning” when submitting your manuscript online. Both the Guide for Authors and the submission portal could be found on the Journal Homepage: Information Fusion | Journal | ScienceDirect.com by Elsevier.

Timeline:

Submission Open Date *08/07/2024
Final Manuscript Submission Deadline *31/12/2024
Editorial Acceptance Deadline *15/04/2025

Keywords:

Artificial Intelligence, Big Data Mining, Ensemble Learning, Mixture of Experts (MoE), Parallel and Distributed Processing, Large Language Model (LLM)

Special Issue on Data Fusion in Modern Energy Systems
截稿日期: 2025-03-15

Modern energy systems encompass a wide range of technologies and infrastructures, including renewable energy sources like solar and wind power, advanced energy storage solutions such as batteries and pumped hydro storage, and smart grids. The integration and coordination of these diverse components present significant challenges in data processing for effective management and operation due to their complexity and variability. Leveraging multi-modal data and employing multi-model approaches, data fusion can merge information from various sensors, measurement devices, and data streams. This enables comprehensive monitoring, improved fault detection, enhanced predictive maintenance, and optimized control strategies for modern energy systems. However, applying the appropriate data fusion strategy to design an advanced energy system is not straightforward. This complexity arises from the need to carefully consider the specific characteristics of the energy system, the types of data involved, and the desired outcomes, making the process highly intricate and context-dependent.

This special issue focuses on data fusion technologies in modern energy systems, aiming to provide the research community with a deeper understanding of data fusion strategies for building advanced energy systems, along with their principles, advantages, and potential applications. By addressing the challenges in the integration and management of these systems, it seeks to demonstrate how cutting-edge data fusion methods can transform the efficiency, reliability, and sustainability of modern energy systems.

Topics of interest include, but are not limited to:

    Innovative data fusion techniques for renewable energy integration
    Real-time data fusion for smart grid management
    Predictive maintenance in energy systems using data fusion
    Multi-modal data fusion for enhanced energy storage management
    Data fusion methods for cyber-physical systems in energy
    Data fusion for fault detection and diagnosis in power systems
    Optimization of energy distribution networks through data fusion
    Data fusion in microgrid management and operation
    Environmental impact assessment using data fusion in energy systems

Guest editors:

Long Cheng, PhD
North China Electric Power University, Beijing, China
(Email: lcheng@ncepu.edu.cn)

Shan Zuo, PhD
University of Connecticut, Storrs, CT, USA
(Email: shan.zuo@uconn.edu)

Pedro P. Vergara, PhD
Delft University of Technology, Delft, Netherlands
(Email: P.P.VergaraBarrios@tudelft.nl)

Tomas Ward, PhD
Dublin City University, Dublin, Ireland
(Email: tomas.ward@dcu.ie)

Xin Ning, PhD
Chinese Academy of Sciences, Beijing, China
(E-mail: ningxin@semi.ac.cn)

Manuscript submission information:

The journal's submission platform (Editorial Manager®) will be available for receiving submissions to this Special Issue from October 1st, 2024. Please refer to the Guide for Authors to prepare your manuscript and select the article type of “VSI: Data Fusion in MES” when submitting your manuscript online. Both the Guide for Authors and the submission portal could be found on the Journal Homepage: Information Fusion | Journal | ScienceDirect.com by Elsevier.

Timeline:

Submission Open Date *01/10/2024
Final Manuscript Submission Deadline *15/03/2025
Editorial Acceptance Deadline *15/10/2025

Keywords:

Data Fusion; Energy System; Machine Learning

Special Issue on GenAI for Information Fusion
截稿日期: 2025-06-30

In recent years, the advancements in Artificial Intelligence (AI), particularly in generative models like ChatGPT and Dall-E, have revolutionized numerous fields including natural language processing, computer vision, and decision-making systems. The convergence of these technologies with information fusion methodologies presents unprecedented opportunities for enhanced data integration, interpretation, and application. However, information fusion methods and applications based on generative AI are still in their nascent stages and have yet to be fully explored. We intend to facilitate innovative applications through this special issue, which focuses on the emerging trends at the intersection of Generative Artificial Intelligence (GenAI) and Information Fusion.

The integration of GenAI into information fusion processes addresses critical challenges in handling diverse data sources, enhancing predictive capabilities, and improving decision-making processes. As data continues to proliferate in volume and complexity, traditional information fusion techniques face limitations in scalability, accuracy, and efficiency. GenAI offers transformative potential to overcome these barriers by generating synthetic data, improving data representation, and enabling more sophisticated fusion algorithms. This special issue seeks to gather pioneering research that demonstrates how GenAI can be effectively utilized in information fusion to address complex problems across various domains such as healthcare, education, cybersecurity, autonomous systems, environmental monitoring, etc.

We invite high-quality submissions that present original research, case studies, and reviews in areas including, but not limited to:

    Theoretical Foundations of GenAI for Information Fusion: New algorithms, models, and frameworks that integrate generative models with information fusion techniques.
    Neural and Symbolic Fusion by GenAI: Methods and fusion mechanisms for representation and reasoning of neural and symbolic technologies based on GenAI.
    Applications of GenAI in Multisource Data Fusion: Case studies and applications in fields such as healthcare, education, cybersecurity, autonomous systems, environmental monitoring, etc.
    Multimodal GenAI: Methods for representation of multimodal data, such visual, text, graph, audio, etc.
    Synthetic Data Generation and Utilization: Methods for creating and using synthetic data to enhance information fusion processes. Detection of GenAI content is also welcomed.
    Performance Evaluation and Benchmarking: Comparative studies on the effectiveness of GenAI in improving the accuracy and efficiency of information fusion.
    Scalability and Real-time Fusion: Techniques for applying GenAI in real-time information fusion scenarios and large-scale data environments.
    Ethical, Legal, and Social Implications: Examination of the ethical considerations, biases, and regulatory aspects associated with the use of GenAI in information fusion.

Guest editors:

Yifan Zhu, PhD
Beijing University of Posts and Telecommunications, Beijing, China.
Email: yifan_zhu@bupt.edu.cn

Kaize Shi, PhD
University of Technology Sydney, Sydney, Australia.
Email: Kaize.Shi@uts.edu.au

Qika Lin, PhD
National University of Singapore,Singapore, Singapore.
Email: linqika@nus.edu.sg

Jie Ma, PhD
Xi’an Jiaotong University, Xi'an, China.
Email: jiema@xjtu.edu.cn

Manuscript submission information:

The journal's submission platform (Editorial Manager®) will be available for receiving submissions to this Special Issue from November 1st, 2024. Please refer to the Guide for Authors to prepare your manuscript and select the article type of “VSI: GenAI for INFFUS” when submitting your manuscript online. Both the Guide for Authors and the submission portal could be found on the Journal Homepage: Information Fusion | Journal | ScienceDirect.com by Elsevier.

Timeline:

Submission Open Date *01/11/2024
Final Manuscript Submission Deadline *30/06/2025
Editorial Acceptance Deadline *01/11/2025

Keywords:

(generative artificial intelligence) OR (generative) AND (AI) OR (generative) AND (fusion) OR (AI) AND (fusion)