Pub Date : 2022-09-26DOI: 10.1109/ISC255366.2022.9922268
Linglong Meng, Stefan Schaffer, Vincent Wappenschmitt
Social group cycling shows a positive impact on facilitating urban cycling as a sustainable means of mobility while increasing cycling safety in urban areas [1]. We present a new urban mobility concept, Connected Swarm Cycling, that creates a group of people cycling together for a while in a common direction or destination. We assume that the concept of Swarm Cycling can significantly change the mobility behaviour of citizens and will be a building block of green mobility for sustainable cities in the future. Utilizing an OSRM11http://project-osrm.org/ routing service with support of trip intersection computing, the system inducts the cyclists into a cycling swarm. The swarms are formed automatically via peer-to-peer connection when cyclists come in proximity, and the information of the swarm and individual cyclist will be synchronized within the swarm via a Nearby Mesh Network. Supporting the implicit interaction within or between swarms, smart wearables are utilized to realize use cases like swarm member identification or signalling in case of merging or splitting of swarms. In this paper, we also present a technical description of our system, including the protocol and network model to support the coordination and synchronization within the swarms.
{"title":"A Connected Swarm Cycling System","authors":"Linglong Meng, Stefan Schaffer, Vincent Wappenschmitt","doi":"10.1109/ISC255366.2022.9922268","DOIUrl":"https://doi.org/10.1109/ISC255366.2022.9922268","url":null,"abstract":"Social group cycling shows a positive impact on facilitating urban cycling as a sustainable means of mobility while increasing cycling safety in urban areas [1]. We present a new urban mobility concept, Connected Swarm Cycling, that creates a group of people cycling together for a while in a common direction or destination. We assume that the concept of Swarm Cycling can significantly change the mobility behaviour of citizens and will be a building block of green mobility for sustainable cities in the future. Utilizing an OSRM11http://project-osrm.org/ routing service with support of trip intersection computing, the system inducts the cyclists into a cycling swarm. The swarms are formed automatically via peer-to-peer connection when cyclists come in proximity, and the information of the swarm and individual cyclist will be synchronized within the swarm via a Nearby Mesh Network. Supporting the implicit interaction within or between swarms, smart wearables are utilized to realize use cases like swarm member identification or signalling in case of merging or splitting of swarms. In this paper, we also present a technical description of our system, including the protocol and network model to support the coordination and synchronization within the swarms.","PeriodicalId":277015,"journal":{"name":"2022 IEEE International Smart Cities Conference (ISC2)","volume":"392 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132655607","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-26DOI: 10.1109/ISC255366.2022.9922507
Phil Aupke, A. Kassler, A. Theocharis, M. Nilsson, Isac Myrén Andersson
Prediction of solar power generation is important in order to optimize energy exchanges in future micro-grids that integrate a large amount of photovoltaics. However, an accurate prediction is difficult due to the uncertainty of weather phenomena that impact produced power. In this paper, we evaluate the impact of different clustering methods on the forecast accuracy for predicting hourly ahead solar power when using machine learning based prediction approaches trained on weather and generated power features. In particular, we compare clustering methods using clearness index and K-means clustering, where we use both euclidian distance and dynamic time-warping. For evaluating prediction accuracy, we develop and compare different prediction models for each of the clusters using production data from a swedish SmartGrid. We demonstrate that proper tuning of thresholds for the clearness index improves prediction accuracy by 20.19% but results in worse performance than using K-means with all weather features as input to the clustering.
{"title":"Impact of Clustering Methods on Machine Learning-based Solar Power Prediction Models","authors":"Phil Aupke, A. Kassler, A. Theocharis, M. Nilsson, Isac Myrén Andersson","doi":"10.1109/ISC255366.2022.9922507","DOIUrl":"https://doi.org/10.1109/ISC255366.2022.9922507","url":null,"abstract":"Prediction of solar power generation is important in order to optimize energy exchanges in future micro-grids that integrate a large amount of photovoltaics. However, an accurate prediction is difficult due to the uncertainty of weather phenomena that impact produced power. In this paper, we evaluate the impact of different clustering methods on the forecast accuracy for predicting hourly ahead solar power when using machine learning based prediction approaches trained on weather and generated power features. In particular, we compare clustering methods using clearness index and K-means clustering, where we use both euclidian distance and dynamic time-warping. For evaluating prediction accuracy, we develop and compare different prediction models for each of the clusters using production data from a swedish SmartGrid. We demonstrate that proper tuning of thresholds for the clearness index improves prediction accuracy by 20.19% but results in worse performance than using K-means with all weather features as input to the clustering.","PeriodicalId":277015,"journal":{"name":"2022 IEEE International Smart Cities Conference (ISC2)","volume":"04 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130700821","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-26DOI: 10.1109/ISC255366.2022.9921897
Amin Mallek, Daniel Klosa, C. Büskens
Short-term flow forecast is a fundamental key in intelligent transportation planning. Often accurate predictions are provided by the predictive models the most adapted to the nature of the addressed problem. In this paper we present a k-Nearest Neighbor approach (E-KNN) enhanced by taking advantage of traffic attributes. The proposed model is applied to 11 weeks of non-processed data, recorded by 7 inductive loop detectors installed on urban roads located in downtown of Bremen (Germany). The performance of E-KNN is tested on 3 weeks of data and reported following different day-hours categories, including rush hours. Excluding early day-hours where traffic is insignificant, E-KNN performs 6-steps (1h) prediction with an average absolute relative error of 17% on test-set.
{"title":"Enhanced K-Nearest Neighbor Model For Multi-steps Traffic Flow Forecast in Urban Roads","authors":"Amin Mallek, Daniel Klosa, C. Büskens","doi":"10.1109/ISC255366.2022.9921897","DOIUrl":"https://doi.org/10.1109/ISC255366.2022.9921897","url":null,"abstract":"Short-term flow forecast is a fundamental key in intelligent transportation planning. Often accurate predictions are provided by the predictive models the most adapted to the nature of the addressed problem. In this paper we present a k-Nearest Neighbor approach (E-KNN) enhanced by taking advantage of traffic attributes. The proposed model is applied to 11 weeks of non-processed data, recorded by 7 inductive loop detectors installed on urban roads located in downtown of Bremen (Germany). The performance of E-KNN is tested on 3 weeks of data and reported following different day-hours categories, including rush hours. Excluding early day-hours where traffic is insignificant, E-KNN performs 6-steps (1h) prediction with an average absolute relative error of 17% on test-set.","PeriodicalId":277015,"journal":{"name":"2022 IEEE International Smart Cities Conference (ISC2)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126924700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-26DOI: 10.1109/ISC255366.2022.9922513
P. Fokaides, A. Jurelionis, Paulius Spudys
At an era when the design of the built environment is being digitised, and the evaluation of buildings is implemented with the use of Industry 4.0 tools, the assessment of the energy performance of buildings can be no exception. Practices such as transmission of information through IoT and digital twins, for the assessment and control of building units, design using Building Information Modelling (BIM), smart meters and digital logbooks are anticipated to be established for conducting the energy assessment of buildings in the near future. Also, additional layers of information related to the sustainability of the built environment have been recently developed, which can enhance the information provided to the building owners and users, regarding the environmental performance of a building. This study presents the overall objectives of the project “Boosting Research for a Smart and Carbon Neutral Built Environment with Digital Twins - SmartWins”, which is funded under the call HORIZON-WIDERA-2021-ACCESS-03 - Twinning. SmartWins project aims to build the capacities for the Kaunas University of Technology in Lithuania, through its “Sustainable Energy in the Built Environment” Research Group (SEBERG) within the Faculty of Civil Engineering and Architecture to conduct high-quality research on the topic of next generation digital twins, applied for allowing the transition to a smart, sustainable, resilient and carbon neutral built environment. The concept of the SmartWins project is to form a network between KTU and leading institutions in the field of energy and sustainability assessment of buildings with the use of Industry 4.0 practices related research and innovation management, for know-how transfer and development of a long-term research collaboration. KTU will twin with the Politecnico di Milano University (PoliMi, Italy), the Centre for Research and Technology, Hellas (CERTH, Greece), a spin-off of the Technical University of Berlin, Contecht GmbH (CON, Germany), and Innotrope (France), aiming to increase its excellence and international reputation in the field, to both cover fundamental research aspects, as well as to further develop its skills, practices and structures to conduct top-notch research.
在建筑环境设计数字化、建筑评估使用工业4.0工具实施的时代，建筑能源性能评估也不例外。预计在不久的将来，将建立诸如通过物联网和数字孪生传输信息，用于建筑单元的评估和控制，使用建筑信息模型(BIM)设计，智能电表和数字日志等实践，以进行建筑物的能源评估。此外，最近还开发了与建筑环境可持续性相关的附加信息层，这可以增强向建筑物所有者和用户提供的有关建筑物环境性能的信息。本研究介绍了“利用数字孪生促进智能和碳中和建筑环境研究- SmartWins”项目的总体目标，该项目由HORIZON-WIDERA-2021-ACCESS-03 -孪生项目资助。SmartWins项目旨在通过其土木工程和建筑学院的“建筑环境中的可持续能源”研究小组(SEBERG)，为立陶宛考纳斯理工大学建立能力，对下一代数字双胞胎主题进行高质量的研究，应用于向智能、可持续、有弹性和碳中和的建筑环境过渡。SmartWins项目的概念是在KTU和建筑能源和可持续性评估领域的领先机构之间建立一个网络，使用工业4.0实践相关的研究和创新管理，以实现技术转让和长期研究合作的发展。KTU将与米兰理工大学(Politecnico di Milano University, Italy)、Hellas研究与技术中心(CERTH, Greece)、柏林技术大学、Contecht GmbH (CON, Germany)和Innotrope (France)合作，旨在提高其在该领域的卓越和国际声誉，涵盖基础研究方面，并进一步发展其技能、实践和结构，以开展一流的研究。
{"title":"Boosting Research for a Smart and Carbon Neutral Built Environment with Digital Twins (SmartWins)","authors":"P. Fokaides, A. Jurelionis, Paulius Spudys","doi":"10.1109/ISC255366.2022.9922513","DOIUrl":"https://doi.org/10.1109/ISC255366.2022.9922513","url":null,"abstract":"At an era when the design of the built environment is being digitised, and the evaluation of buildings is implemented with the use of Industry 4.0 tools, the assessment of the energy performance of buildings can be no exception. Practices such as transmission of information through IoT and digital twins, for the assessment and control of building units, design using Building Information Modelling (BIM), smart meters and digital logbooks are anticipated to be established for conducting the energy assessment of buildings in the near future. Also, additional layers of information related to the sustainability of the built environment have been recently developed, which can enhance the information provided to the building owners and users, regarding the environmental performance of a building. This study presents the overall objectives of the project “Boosting Research for a Smart and Carbon Neutral Built Environment with Digital Twins - SmartWins”, which is funded under the call HORIZON-WIDERA-2021-ACCESS-03 - Twinning. SmartWins project aims to build the capacities for the Kaunas University of Technology in Lithuania, through its “Sustainable Energy in the Built Environment” Research Group (SEBERG) within the Faculty of Civil Engineering and Architecture to conduct high-quality research on the topic of next generation digital twins, applied for allowing the transition to a smart, sustainable, resilient and carbon neutral built environment. The concept of the SmartWins project is to form a network between KTU and leading institutions in the field of energy and sustainability assessment of buildings with the use of Industry 4.0 practices related research and innovation management, for know-how transfer and development of a long-term research collaboration. KTU will twin with the Politecnico di Milano University (PoliMi, Italy), the Centre for Research and Technology, Hellas (CERTH, Greece), a spin-off of the Technical University of Berlin, Contecht GmbH (CON, Germany), and Innotrope (France), aiming to increase its excellence and international reputation in the field, to both cover fundamental research aspects, as well as to further develop its skills, practices and structures to conduct top-notch research.","PeriodicalId":277015,"journal":{"name":"2022 IEEE International Smart Cities Conference (ISC2)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124300491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-26DOI: 10.1109/ISC255366.2022.9922561
Arthur Souza, N. Cacho, T. Batista
The increase in the computing capabilities of Edge devices made it possible to distribute the processing contracted in the Cloud, leveraging the emergence of Edge Computing and Fog Computing. Fog's improved processing of data obtained by Edge quickly progressed from simple cleaning and categorization to more refined and contextually related information. Thus, there is a growing need for persistent storage at the Fog/Edge level, especially in facing the scenarios present in Osmotic Computing. With this context in mind, our work presents a solution for data persistence between the various levels of Edge/Fog/Cloud. Going further, we introduce Fogmotic, a Database as a Service platform that focuses on distribution, synchronization, reliability, efficiency, and data migration at the Edge/Fog/Cloud levels. Finally, we present an experimental evaluation of the reading, writing, and sync rate performance obtained by Fogmotic.
{"title":"Fogmotic: Applying Osmotic Data Services to improve Database Operations on SmartCity Environments","authors":"Arthur Souza, N. Cacho, T. Batista","doi":"10.1109/ISC255366.2022.9922561","DOIUrl":"https://doi.org/10.1109/ISC255366.2022.9922561","url":null,"abstract":"The increase in the computing capabilities of Edge devices made it possible to distribute the processing contracted in the Cloud, leveraging the emergence of Edge Computing and Fog Computing. Fog's improved processing of data obtained by Edge quickly progressed from simple cleaning and categorization to more refined and contextually related information. Thus, there is a growing need for persistent storage at the Fog/Edge level, especially in facing the scenarios present in Osmotic Computing. With this context in mind, our work presents a solution for data persistence between the various levels of Edge/Fog/Cloud. Going further, we introduce Fogmotic, a Database as a Service platform that focuses on distribution, synchronization, reliability, efficiency, and data migration at the Edge/Fog/Cloud levels. Finally, we present an experimental evaluation of the reading, writing, and sync rate performance obtained by Fogmotic.","PeriodicalId":277015,"journal":{"name":"2022 IEEE International Smart Cities Conference (ISC2)","volume":"152 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122149995","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-26DOI: 10.1109/ISC255366.2022.9922599
Hajer Alyammahi, P. Liatsis
Providing ancillary services for future smart grids is challenging because of the rapidly growing electricity demand, while having uncertainties in renewable power generation, limited availability of conventional spinning reserves, and expensive storage systems. Thus, Home Energy Management Systems (HEMSs) have been gaining increased attention nowadays. To capitalize on the potential of HEMS, which supports customer participation and two-way power communication so as to maintain the generation-load balance, two interconnected challenges, i.e., load monitoring and identification of appliances consumption, need to be addressed. In this contribution, a comprehensive nonintrusive load monitoring (NILM) algorithm for appliance identification is proposed, which only requires a single sensing point from conventional homes, i.e., the aggregated power signal. Machine learning algorithms and both time-domain and frequency-domain based feature extraction are utilized in the development of the proposed solution. Simulation experiments are performed using the Reference Energy Disaggregation Dataset (REDD), a real household power consumption dataset. Simulation results demonstrate the effectiveness of the proposed NILM strategy with F1-score values of 97.659%, higher than those reported in the state-of-the-art.
{"title":"Towards Identification of Appliances in Conventional Homes using ML and Descriptive Statistics","authors":"Hajer Alyammahi, P. Liatsis","doi":"10.1109/ISC255366.2022.9922599","DOIUrl":"https://doi.org/10.1109/ISC255366.2022.9922599","url":null,"abstract":"Providing ancillary services for future smart grids is challenging because of the rapidly growing electricity demand, while having uncertainties in renewable power generation, limited availability of conventional spinning reserves, and expensive storage systems. Thus, Home Energy Management Systems (HEMSs) have been gaining increased attention nowadays. To capitalize on the potential of HEMS, which supports customer participation and two-way power communication so as to maintain the generation-load balance, two interconnected challenges, i.e., load monitoring and identification of appliances consumption, need to be addressed. In this contribution, a comprehensive nonintrusive load monitoring (NILM) algorithm for appliance identification is proposed, which only requires a single sensing point from conventional homes, i.e., the aggregated power signal. Machine learning algorithms and both time-domain and frequency-domain based feature extraction are utilized in the development of the proposed solution. Simulation experiments are performed using the Reference Energy Disaggregation Dataset (REDD), a real household power consumption dataset. Simulation results demonstrate the effectiveness of the proposed NILM strategy with F1-score values of 97.659%, higher than those reported in the state-of-the-art.","PeriodicalId":277015,"journal":{"name":"2022 IEEE International Smart Cities Conference (ISC2)","volume":"312 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122803903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-26DOI: 10.1109/ISC255366.2022.9921921
Muhammad Taimoor Khan
Digital twin-based modern smart city infrastructures are evolving into intelligent and distributed systems of autonomous entities operating in a dynamic cyber-physical environment to offer real-time and critical services. These services are typically implemented as software applications in various application domains, e.g., healthcare, cooperative robotic systems, and autonomous vehicles. However, to assure continued safe operations of the critical services with strict real-time requirements even when the service is under attack is an extremely challenging task mainly because the underlying operating environment for such applications is highly volatile yet distributed. To this end, first, we classify (as we call it) timed resilience requirements into computational and communication resilience and then discuss key challenges that hinder the modeling of such requirements to help develop rigorous distributed applications for real-time resilient autonomous systems. Finally, we demonstrate our vision to handle these challenges by introducing by-design and by-response approaches that consider security as a prerequisite of the safety and resilience of autonomous systems.
{"title":"Challenges in Modelling Applications for Safe and Resilient Digital Twins","authors":"Muhammad Taimoor Khan","doi":"10.1109/ISC255366.2022.9921921","DOIUrl":"https://doi.org/10.1109/ISC255366.2022.9921921","url":null,"abstract":"Digital twin-based modern smart city infrastructures are evolving into intelligent and distributed systems of autonomous entities operating in a dynamic cyber-physical environment to offer real-time and critical services. These services are typically implemented as software applications in various application domains, e.g., healthcare, cooperative robotic systems, and autonomous vehicles. However, to assure continued safe operations of the critical services with strict real-time requirements even when the service is under attack is an extremely challenging task mainly because the underlying operating environment for such applications is highly volatile yet distributed. To this end, first, we classify (as we call it) timed resilience requirements into computational and communication resilience and then discuss key challenges that hinder the modeling of such requirements to help develop rigorous distributed applications for real-time resilient autonomous systems. Finally, we demonstrate our vision to handle these challenges by introducing by-design and by-response approaches that consider security as a prerequisite of the safety and resilience of autonomous systems.","PeriodicalId":277015,"journal":{"name":"2022 IEEE International Smart Cities Conference (ISC2)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124902530","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-26DOI: 10.1109/ISC255366.2022.9921957
Ivo S. L. Tebexreni, Carmen L. T. Borges
This article proposes methods that use nonsequential Monte Carlo Simulation (MCS) to calculate reliability indices of power systems with correlated energy sources. The methods apply Principal Correlated Analysis (PCA), covariance matrix, random variable transformation and correlation mapping. Good results were found in cases with linear correlations and high failure state frequency. The processing time was consistent with that observed in classical nonsequential Monte Carlo simulation, and with PCA, it was possible to reduce the dimensionality of the system.
{"title":"Efficient Methods to Calculate the Reliability of Energy Systems with Correlated Renewable Sources","authors":"Ivo S. L. Tebexreni, Carmen L. T. Borges","doi":"10.1109/ISC255366.2022.9921957","DOIUrl":"https://doi.org/10.1109/ISC255366.2022.9921957","url":null,"abstract":"This article proposes methods that use nonsequential Monte Carlo Simulation (MCS) to calculate reliability indices of power systems with correlated energy sources. The methods apply Principal Correlated Analysis (PCA), covariance matrix, random variable transformation and correlation mapping. Good results were found in cases with linear correlations and high failure state frequency. The processing time was consistent with that observed in classical nonsequential Monte Carlo simulation, and with PCA, it was possible to reduce the dimensionality of the system.","PeriodicalId":277015,"journal":{"name":"2022 IEEE International Smart Cities Conference (ISC2)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123353552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-09-26DOI: 10.1109/ISC255366.2022.9921824
Tanveer Hussain, Hang Dai, W. Gueaieb, Marco Sicklinger, Giulia De Masi
Effective fire detection using vision sensors is a widely accepted challenge in smart cities and rural areas, where forest and building fires significantly contribute to the loss of human lives and properties. Early fire detection using deep learning techniques is emerged to be an effective solution using close-circuit television (CCTV) in smart cities, but it has limited coverage in huge building infrastructures and urban forests. Unmanned Aerial Vehicles (UAV) cover wide areas, but fire detection in visual data captured from UAVs is a challenging task. Therefore, we employ deep multi-scale features from a backbone model and apply attention mechanism for accurate fire detection. The deep features from intermediate layers capture fire regions using spatial object edges information and final layers extract image global representations. The features fusion ensures to represent the image effectively, where the fused features are enhanced using multi-headed self-attention to highlight the most important fire regions. Preliminary experimental results (https://github.com/tanveer-hussain/DMFA-Fire) using UAV fire detection dataset demonstrate effective performance of the proposed model against rivals and consequently present a new deep model's perspective to consider multi layer features for accurate detection performance, thereby providing effective applicability in smart cities environments.
{"title":"UAV-based Multi-scale Features Fusion Attention for Fire Detection in Smart City Ecosystems","authors":"Tanveer Hussain, Hang Dai, W. Gueaieb, Marco Sicklinger, Giulia De Masi","doi":"10.1109/ISC255366.2022.9921824","DOIUrl":"https://doi.org/10.1109/ISC255366.2022.9921824","url":null,"abstract":"Effective fire detection using vision sensors is a widely accepted challenge in smart cities and rural areas, where forest and building fires significantly contribute to the loss of human lives and properties. Early fire detection using deep learning techniques is emerged to be an effective solution using close-circuit television (CCTV) in smart cities, but it has limited coverage in huge building infrastructures and urban forests. Unmanned Aerial Vehicles (UAV) cover wide areas, but fire detection in visual data captured from UAVs is a challenging task. Therefore, we employ deep multi-scale features from a backbone model and apply attention mechanism for accurate fire detection. The deep features from intermediate layers capture fire regions using spatial object edges information and final layers extract image global representations. The features fusion ensures to represent the image effectively, where the fused features are enhanced using multi-headed self-attention to highlight the most important fire regions. Preliminary experimental results (https://github.com/tanveer-hussain/DMFA-Fire) using UAV fire detection dataset demonstrate effective performance of the proposed model against rivals and consequently present a new deep model's perspective to consider multi layer features for accurate detection performance, thereby providing effective applicability in smart cities environments.","PeriodicalId":277015,"journal":{"name":"2022 IEEE International Smart Cities Conference (ISC2)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129578228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：