https://www.mdu.se/

mdu.sePublikationer
Ändra sökning
Länk till posten
Permanent länk

Direktlänk
Publikationer (10 of 101) Visa alla publikationer
Seceleanu, T., Xiong, N., Enoiu, E. P. & Seceleanu, C. (2024). Building a Digital Twin Framework for Dynamic and Robust Distributed Systems. In: Lect. Notes Comput. Sci.: . Paper presented at 8th International Conference on Engineering of Computer-Based Systems, ECBS 2023, Västerås, 16 October 2023 through 18 October 2023 (pp. 254-258). Springer Science and Business Media Deutschland GmbH
Öppna denna publikation i ny flik eller fönster >>Building a Digital Twin Framework for Dynamic and Robust Distributed Systems
2024 (Engelska)Ingår i: Lect. Notes Comput. Sci., Springer Science and Business Media Deutschland GmbH , 2024, s. 254-258Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

Digital Twins (DTs) serve as the backbone of Industry 4.0, offering virtual representations of actual systems, enabling accurate simulations, analysis, and control. These representations help in predicting system behaviour, facilitating multiple real-time tests, and reducing risks and costs while identifying optimization areas. DTs meld cyber and physical realms, accelerating the design and modelling of sustainable innovations. Despite their potential, the complexity of DTs presents challenges in their industrial application. We sketch here an approach to build an adaptable and trustable framework for building and operating DT systems, which is the basis for the academia-industry project A Digital Twin Framework for Dynamic and Robust Distributed Systems (D-RODS). D-RODS aims to address the challenges above, aiming to advance industrial digitalization and targeting areas like system efficiency, incorporating AI and verification techniques with formal support. 

Ort, förlag, år, upplaga, sidor
Springer Science and Business Media Deutschland GmbH, 2024
Serie
Lecture Notes in Computer Science, ISSN 0302-9743 ; 14390 LNCS
Nyckelord
AI, digital twins, industrial automation, resource utilization, verification and validation, Actual system, Analysis and controls, Distributed systems, Resources utilizations, Simulation analysis, Simulation control, System behaviors, Verification-and-validation, Virtual representations, Artificial intelligence
Nationell ämneskategori
Datavetenskap (datalogi)
Identifikatorer
urn:nbn:se:mdh:diva-65247 (URN)10.1007/978-3-031-49252-5_22 (DOI)2-s2.0-85180149728 (Scopus ID)9783031492518 (ISBN)
Konferens
8th International Conference on Engineering of Computer-Based Systems, ECBS 2023, Västerås, 16 October 2023 through 18 October 2023
Tillgänglig från: 2024-01-03 Skapad: 2024-01-03 Senast uppdaterad: 2024-01-03Bibliografiskt granskad
Zhang, L., Xiong, N., Gao, W. & Wu, P. (2024). Improved Detection Method for Micro-Targets in Remote Sensing Images. Information, 15(2), Article ID 108.
Öppna denna publikation i ny flik eller fönster >>Improved Detection Method for Micro-Targets in Remote Sensing Images
2024 (Engelska)Ingår i: Information, E-ISSN 2078-2489, Vol. 15, nr 2, artikel-id 108Artikel i tidskrift (Refereegranskat) Published
Abstract [en]

With the exponential growth of remote sensing images in recent years, there has been a significant increase in demand for micro-target detection. Recently, effective detection methods for small targets have emerged; however, for micro-targets (even fewer pixels than small targets), most existing methods are not fully competent in feature extraction, target positioning, and rapid classification. This study proposes an enhanced detection method, especially for micro-targets, in which a combined loss function (consisting of NWD and CIOU) is used instead of a singular CIOU loss function. In addition, the lightweight Content-Aware Reassembly of Features (CARAFE) replaces the original bilinear interpolation upsampling algorithm, and a spatial pyramid structure is added into the network model’s small target layer. The proposed algorithm undergoes training and validation utilizing the benchmark dataset known as AI-TOD. Compared to speed-oriented YOLOv7-tiny, the mAP0.5 and mAP0.5:0.95 of our improved algorithm increased from 42.0% and 16.8% to 48.7% and 18.9%, representing improvements of 6.7% and 2.1%, respectively, while the detection speed was almost equal to that of YOLOv7-tiny. Furthermore, our method was also tested on a dataset of multi-scale targets, which contains small targets, medium targets, and large targets. The results demonstrated that mAP0.5:0.95 increased from “9.8%, 54.8%, and 68.2%” to “12.6%, 55.6%, and 70.1%” for detection across different scales, indicating improvements of 2.8%, 0.8%, and 1.9%, respectively. In summary, the presented method improves detection metrics for micro-targets in various scenarios while satisfying the requirements of detection speed in a real-time system.

Ort, förlag, år, upplaga, sidor
Multidisciplinary Digital Publishing Institute (MDPI), 2024
Nyckelord
CARAFE, micro-targets, NWD, remote sensing images, spatial pyramid, Feature extraction, Interactive computer systems, Large datasets, Real time systems, Remote sensing, Content-aware, Content-aware reassembly of feature, Detection methods, Loss functions, Micro-target, Reassembly, Small targets, Spatial pyramids, Image enhancement
Nationell ämneskategori
Data- och informationsvetenskap
Identifikatorer
urn:nbn:se:mdh:diva-66179 (URN)10.3390/info15020108 (DOI)001172402600001 ()2-s2.0-85185707305 (Scopus ID)
Anmärkning

Article; Export Date: 06 March 2024; Cited By: 0; Correspondence Address: P. Wu; School of Computer and Information Technology, Beijing Jiaotong University, Beijing, 100044, China; email: 14112078@bjtu.edu.cn

Tillgänglig från: 2024-03-06 Skapad: 2024-03-06 Senast uppdaterad: 2024-03-13Bibliografiskt granskad
Zhang, L., Xiong, N., Pan, X., Yue, X., Wu, P. & Guo, C. (2023). Improved Object Detection Method Utilizing YOLOv7-Tiny for Unmanned Aerial Vehicle Photographic Imagery. Algorithms, 16(11), Article ID 520.
Öppna denna publikation i ny flik eller fönster >>Improved Object Detection Method Utilizing YOLOv7-Tiny for Unmanned Aerial Vehicle Photographic Imagery
Visa övriga...
2023 (Engelska)Ingår i: Algorithms, E-ISSN 1999-4893, Vol. 16, nr 11, artikel-id 520Artikel i tidskrift (Refereegranskat) Published
Abstract [en]

In unmanned aerial vehicle photographs, object detection algorithms encounter challenges in enhancing both speed and accuracy for objects of different sizes, primarily due to complex backgrounds and small objects. This study introduces the PDWT-YOLO algorithm, based on the YOLOv7-tiny model, to improve the effectiveness of object detection across all sizes. The proposed method enhances the detection of small objects by incorporating a dedicated small-object detection layer, while reducing the conflict between classification and regression tasks through the replacement of the YOLOv7-tiny model’s detection head (IDetect) with a decoupled head. Moreover, network convergence is accelerated, and regression accuracy is improved by replacing the Complete Intersection over Union (CIoU) loss function with a Wise Intersection over Union (WIoU) focusing mechanism in the loss function. To assess the proposed model’s effectiveness, it was trained and tested on the VisDrone-2019 dataset comprising images captured by various drones across diverse scenarios, weather conditions, and lighting conditions. The experiments show that mAP@0.5:0.95 and mAP@0.5 increased by 5% and 6.7%, respectively, with acceptable running speed compared with the original YOLOv7-tiny model. Furthermore, this method shows improvement over other datasets, confirming that PDWT-YOLO is effective for multiscale object detection. 

Ort, förlag, år, upplaga, sidor
Multidisciplinary Digital Publishing Institute (MDPI), 2023
Nyckelord
decoupled head, small-object detection, WIoU, YOLOv7-tiny model, Aerial photography, Antennas, Image enhancement, Object recognition, Unmanned aerial vehicles (UAV), Aerial vehicle, Loss functions, Object detection method, Objects detection, Photographic imagery, Small object detection, Small objects, Wise intersection over union, Object detection
Nationell ämneskategori
Datorseende och robotik (autonoma system)
Identifikatorer
urn:nbn:se:mdh:diva-65015 (URN)10.3390/a16110520 (DOI)001115260900001 ()2-s2.0-85178308460 (Scopus ID)
Tillgänglig från: 2023-12-13 Skapad: 2023-12-13 Senast uppdaterad: 2023-12-20Bibliografiskt granskad
Wu, P., Xiong, N., Li, G. & lv, J. (2023). Incremental Bayesian Classifier for Streaming Data with Concept Drift. In: Lecture Notes on Data Engineering and Communications Technologies: (pp. 509-518). Springer Science and Business Media Deutschland GmbH, 153
Öppna denna publikation i ny flik eller fönster >>Incremental Bayesian Classifier for Streaming Data with Concept Drift
2023 (Engelska)Ingår i: Lecture Notes on Data Engineering and Communications Technologies, Springer Science and Business Media Deutschland GmbH , 2023, Vol. 153, s. 509-518Kapitel i bok, del av antologi (Övrigt vetenskapligt)
Abstract [en]

Classification is an important task in the field of machine learning. Most classifiers based on offline learning are invalid for open data streams. In contrast, incremental learning is feasible for continuous data. This paper presents the Incremental Bayesian Classifier “Incremental_BC”, which continuously updates the probabilistic information according to each new training sample via recursive calculation. Further, the Incremental_BC is improved to deal with the flowing data whose distribution and property evolve with time, i.e., the concept drift. The effectiveness of the proposed methods has been verified by the results of simulation tests on benchmark data sets.

Ort, förlag, år, upplaga, sidor
Springer Science and Business Media Deutschland GmbH, 2023
Nyckelord
Classification (of information), Data streams, E-learning, Bayesian classifier, Concept drifts, Continuous data, Data stream, Incremental learning, Machine-learning, Off-line learning, Online learning, Open datum, Streaming data, Open Data, Concept drift
Nationell ämneskategori
Datavetenskap (datalogi)
Identifikatorer
urn:nbn:se:mdh:diva-61963 (URN)10.1007/978-3-031-20738-9_58 (DOI)000964184200058 ()2-s2.0-85147842399 (Scopus ID)
Tillgänglig från: 2023-02-22 Skapad: 2023-02-22 Senast uppdaterad: 2023-05-03Bibliografiskt granskad
Antonic, N., Khalid, A. H., Hamila, M. E. & Xiong, N. (2023). Online Tuning of PID Controllers Based on Membrane Neural Computing. In: Lecture Notes on Data Engineering and Communications Technologies: (pp. 455-464). Springer Science and Business Media Deutschland GmbH, 153
Öppna denna publikation i ny flik eller fönster >>Online Tuning of PID Controllers Based on Membrane Neural Computing
2023 (Engelska)Ingår i: Lecture Notes on Data Engineering and Communications Technologies, Springer Science and Business Media Deutschland GmbH , 2023, Vol. 153, s. 455-464Kapitel i bok, del av antologi (Övrigt vetenskapligt)
Abstract [en]

PID controllers are still popular in a wide range of engineering practices due to their simplicity and robustness. Traditional design of a PID controller needs manual setting of its parameters in advance. This paper proposes a new method for online tuning of PID controllers based on hybridized neural membrane computing. A neural network is employed to adaptively determine the proper values of the PID parameters in terms of evolving situations/stages in the control process. Further the learning of the neural network is performed based on a membrane algorithm, which is used to locate the weights of the network to optimize the control performance. The effectiveness of the proposed method has been demonstrated by the preliminary results from simulation tests.

Ort, förlag, år, upplaga, sidor
Springer Science and Business Media Deutschland GmbH, 2023
Nyckelord
Bioinformatics, Controllers, Electric control equipment, Membranes, Proportional control systems, Tuning, Engineering practices, Gain tuning, Membrane algorithm, Membrane computing, Neural computing, Neural membranes, Neural-networks, Online gain tuning, Online tuning, PID controllers, Three term control systems, Neural network, PID controller
Nationell ämneskategori
Reglerteknik
Identifikatorer
urn:nbn:se:mdh:diva-61961 (URN)10.1007/978-3-031-20738-9_52 (DOI)000964184200052 ()2-s2.0-85147852132 (Scopus ID)
Tillgänglig från: 2023-02-22 Skapad: 2023-02-22 Senast uppdaterad: 2024-01-18Bibliografiskt granskad
Xiong, N., Li, M., Li, K., Xiao, Z., Liao, L. & Wang, L. (2023). Preface. Lecture Notes on Data Engineering and Communications Technologies, 153, xiii-xiv
Öppna denna publikation i ny flik eller fönster >>Preface
Visa övriga...
2023 (Engelska)Ingår i: Lecture Notes on Data Engineering and Communications Technologies, ISSN 2367-4512, Vol. 153, s. xiii-xivArtikel i tidskrift, Editorial material (Refereegranskat) Published
Ort, förlag, år, upplaga, sidor
Springer Science and Business Media Deutschland GmbH, 2023
Nationell ämneskategori
Datorteknik
Identifikatorer
urn:nbn:se:mdh:diva-61959 (URN)2-s2.0-85147879121 (Scopus ID)
Tillgänglig från: 2023-02-22 Skapad: 2023-02-22 Senast uppdaterad: 2023-02-22Bibliografiskt granskad
Xiong, N. & Punnekkat, S. (2023). Tiny Federated Learning with Bayesian Classifiers. In: IEEE Int Symp Ind Electron: . Paper presented at IEEE International Symposium on Industrial Electronics. Institute of Electrical and Electronics Engineers Inc.
Öppna denna publikation i ny flik eller fönster >>Tiny Federated Learning with Bayesian Classifiers
2023 (Engelska)Ingår i: IEEE Int Symp Ind Electron, Institute of Electrical and Electronics Engineers Inc. , 2023Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

Tiny machine learning (TinyML) represents an emerging research direction that aims to realize machine learning on Internet of Things (IoT) devices. The current TinyML research seems to focus on supporting the deployment of deep learning models on microprocessors, while the models themselves are trained on high performance computers or clouds. However, in the resource/time constrained IoT contexts, it is more desirable to perform data analytics and learning tasks directly on edge devices for crucial benefits such as increased energy efficiency, reduced latency as well as lower communication cost.To address the above challenge, this paper proposes a tiny federated learning algorithm for enabling learning of Bayesian classifiers based on distributed tiny data storage, referred to as TFL-BC. In TFL-BC, Bayesian learning is executed in parallel across multiple edge devices using local (tiny) training data and subsequently the learning results from local devices are aggregated via a central node to obtain the final classification model. The results of experiments conducted on a set of benchmark datasets demonstrate that our algorithm can produce final aggregated models that outperform single tiny Bayesian classifiers and that the result of tiny federated learning (of Bayesian classifier) is independent of the number of data partitions used for generating the distributed local training data.

Ort, förlag, år, upplaga, sidor
Institute of Electrical and Electronics Engineers Inc., 2023
Nyckelord
Bayesian classifier, edge computing, federated learning, tiny machine learning, Classification (of information), Data Analytics, Deep learning, Digital storage, Energy efficiency, Learning algorithms, Learning systems, 'current, High performance computers, Learning models, Machine learning research, Machine-learning, Training data, Internet of things
Nationell ämneskategori
Data- och informationsvetenskap
Identifikatorer
urn:nbn:se:mdh:diva-64430 (URN)10.1109/ISIE51358.2023.10228115 (DOI)2-s2.0-85172110028 (Scopus ID)9798350399714 (ISBN)
Konferens
IEEE International Symposium on Industrial Electronics
Tillgänglig från: 2023-10-09 Skapad: 2023-10-09 Senast uppdaterad: 2023-10-09Bibliografiskt granskad
López, D., Ramírez-Gallego, S., García, S., Xiong, N. & Herrera, F. (2021). BELIEF: A distance-based redundancy-proof feature selection method for Big Data. Information Sciences, 558, 124-139
Öppna denna publikation i ny flik eller fönster >>BELIEF: A distance-based redundancy-proof feature selection method for Big Data
Visa övriga...
2021 (Engelska)Ingår i: Information Sciences, ISSN 0020-0255, E-ISSN 1872-6291, Vol. 558, s. 124-139Artikel i tidskrift (Refereegranskat) Published
Abstract [en]

With the advent of Big Data era, data reduction methods are in highly demand given their ability to simplify huge data, and ease complex learning processes. Concretely, algorithms able to select relevant dimensions from a set of millions are of huge importance. Although effective, these techniques also suffer from the “scalability” curse when they are brought into tackle large-scale problems. In this paper, we propose a distributed feature weighting algorithm which precisely estimates feature importance in large datasets using the well-know algorithm RELIEF in small problems. Our solution, called BELIEF, incorporates a novel redundancy elimination measure that generates similar schemes to those based on entropy, but at a much lower time cost. Furthermore, BELIEF provides a smooth scale-up when more instances are required to increase precision in estimations. Empirical tests performed on our method illustrate the estimation ability of BELIEF in manifold huge sets – both in number of features and instances, as well as its reduced runtime cost as compared to other state-of-the-art methods. 

Ort, förlag, år, upplaga, sidor
Elsevier Inc., 2021
Nyckelord
Apache spark, Big Data, Feature selection (FS), High-dimensional, Redundancy elimination
Nationell ämneskategori
Datavetenskap (datalogi)
Identifikatorer
urn:nbn:se:mdh:diva-53524 (URN)10.1016/j.ins.2020.12.082 (DOI)000634824100008 ()2-s2.0-85100519874 (Scopus ID)
Tillgänglig från: 2021-04-01 Skapad: 2021-04-01 Senast uppdaterad: 2022-08-29Bibliografiskt granskad
Seceleanu, T., Xiong, N. & Seceleanu, C. (2021). Control as a service - Intelligent networking. In: Proceedings - 2021 IEEE 45th Annual Computers, Software, and Applications Conference, COMPSAC 2021: . Paper presented at 45th IEEE Annual Computers, Software, and Applications Conference, COMPSAC 2021, 12 July 2021 through 16 July 2021 (pp. 1887-1892). Institute of Electrical and Electronics Engineers Inc.
Öppna denna publikation i ny flik eller fönster >>Control as a service - Intelligent networking
2021 (Engelska)Ingår i: Proceedings - 2021 IEEE 45th Annual Computers, Software, and Applications Conference, COMPSAC 2021, Institute of Electrical and Electronics Engineers Inc. , 2021, s. 1887-1892Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

The paper introduces elements of a service based perspective of a scalable and dynamic automation system architecture. The approach is based on potentially multi-role devices (implementing node management, processing and networking functionalities) hosting a set of services requested by input nodes. In addition, artificial intelligence support is described to provide means of reaching deployment optimality and reliability. Formal approaches are deemed necessary for both verification of the artificial intelligence approach and of the resulting solutions. A model-based design path is complementary considered in order to lead to an increased efficiency in resource utilization, to lowering design efforts, and ensure a formally correct allocation of services, according to system requirements and constraints.

Ort, förlag, år, upplaga, sidor
Institute of Electrical and Electronics Engineers Inc., 2021
Nyckelord
Dynamic systems, Networked automation systems, Service oriented architectures, Application programs, Automation, Intelligent networks, Automation systems, Design effort, Formal approach, Intelligent networking, Model- based designs, Node management, Resource utilizations, System requirements, Artificial intelligence
Nationell ämneskategori
Data- och informationsvetenskap
Identifikatorer
urn:nbn:se:mdh:diva-56129 (URN)10.1109/COMPSAC51774.2021.00285 (DOI)000706529000274 ()2-s2.0-85115884209 (Scopus ID)9781665424639 (ISBN)
Konferens
45th IEEE Annual Computers, Software, and Applications Conference, COMPSAC 2021, 12 July 2021 through 16 July 2021
Tillgänglig från: 2021-10-07 Skapad: 2021-10-07 Senast uppdaterad: 2021-11-11Bibliografiskt granskad
Dust, L., Murcia, M. L., Mäkilä, A., Nordin, P., Xiong, N. & Herrera, F. (2021). Federated Fuzzy Learning with Imbalanced Data. In: Proceedings - 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021: . Paper presented at 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021 (pp. 1130-1137). Institute of Electrical and Electronics Engineers Inc.
Öppna denna publikation i ny flik eller fönster >>Federated Fuzzy Learning with Imbalanced Data
Visa övriga...
2021 (Engelska)Ingår i: Proceedings - 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021, Institute of Electrical and Electronics Engineers Inc. , 2021, s. 1130-1137Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

Federated learning (FL) is an emerging and privacy-preserving machine learning technique that is shown to be increasingly important in the digital age. The two challenging issues for FL are: (1) communication overhead between clients and the server, and (2) volatile distribution of training data such as class imbalance. The paper aims to tackle these two challenges with the proposal of a federated fuzzy learning algorithm (FFLA) that can be used for data-based construction of fuzzy classification models in a distributed setting. The proposed learning algorithm is fast and highly cheap in communication by requiring only two rounds of interplay between the server and clients. Moreover, FFLA is empowered with an an imbalance adaptation mechanism so that it remains robust against heterogeneous distributions of data and class imbalance. The efficacy of the proposed learning method has been verified by the simulation tests made on a set of balanced and imbalanced benchmark data sets.

Ort, förlag, år, upplaga, sidor
Institute of Electrical and Electronics Engineers Inc., 2021
Nyckelord
Class imbalance, Communication cost, Distributed learning, Federated learning, Fuzzy rule-based model, Imbalanced data, Incremental learning
Nationell ämneskategori
Data- och informationsvetenskap
Identifikatorer
urn:nbn:se:mdh:diva-57626 (URN)10.1109/ICMLA52953.2021.00185 (DOI)000779208200177 ()2-s2.0-85125878050 (Scopus ID)9781665443371 (ISBN)
Konferens
20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021
Tillgänglig från: 2022-03-16 Skapad: 2022-03-16 Senast uppdaterad: 2022-11-08Bibliografiskt granskad
Organisationer
Identifikatorer
ORCID-id: ORCID iD iconorcid.org/0000-0001-9857-4317

Sök vidare i DiVA

Visa alla publikationer