Pub Date : 2023-01-05DOI: 10.1109/IDCIoT56793.2023.10053529
Chinni Roshini Durga, J. Karthik, Reddi Dakshayani, Songa Manikanta
Many fire accidents that occur recently are caused by electric vehicles like Ola, Okinawa, and Pure EV. The primary reason behind catching fire for EV scooters is a thermal runaway, and this occurs due to multiple reasons--melting of electrolyte, operational Temperatures of the battery, poor quality of the battery cells and battery packs, and lack of active cell assemblies. So, the proposed system intends to provide safety for EV users. For this, the battery’s temperature, battery’s size, and the voltage fluctuations of the battery are monitored by fixing the sensors to the battery of the electric vehicle along with other required IoT components. These values are monitored so that whenever these values cross their threshold limit the user will get a notification to mobile phone through mobile application as an alert message. Along with this, temperature readings, size indication, battery’s voltage percentage are also displayed on the display which is connected to the dashboard of the vehicle and also an alarm sound that alerts the user.
{"title":"A Novel Approach for Smart Battery Monitoring System in Electric Vehicles using Internet of Things","authors":"Chinni Roshini Durga, J. Karthik, Reddi Dakshayani, Songa Manikanta","doi":"10.1109/IDCIoT56793.2023.10053529","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053529","url":null,"abstract":"Many fire accidents that occur recently are caused by electric vehicles like Ola, Okinawa, and Pure EV. The primary reason behind catching fire for EV scooters is a thermal runaway, and this occurs due to multiple reasons--melting of electrolyte, operational Temperatures of the battery, poor quality of the battery cells and battery packs, and lack of active cell assemblies. So, the proposed system intends to provide safety for EV users. For this, the battery’s temperature, battery’s size, and the voltage fluctuations of the battery are monitored by fixing the sensors to the battery of the electric vehicle along with other required IoT components. These values are monitored so that whenever these values cross their threshold limit the user will get a notification to mobile phone through mobile application as an alert message. Along with this, temperature readings, size indication, battery’s voltage percentage are also displayed on the display which is connected to the dashboard of the vehicle and also an alarm sound that alerts the user.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"18 1","pages":"870-874"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75022979","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 : 2023-01-05DOI: 10.1109/IDCIoT56793.2023.10053551
Lohitha B, Adithya V, Yasaswi Aparna N, H. R., Srithar S, Aravinth S S
Because of the elevated risk of illness and fatality, chronic renal disease is regarded as a serious health issue. Renal disease is also called kidney disease. Kidney infections are particularly challenging to diagnose since they progress slowly and continuously. For the same reason, a lot of patients wait until the very end stage to diagnose their condition. It’s critical to have trustworthy methods in the early stage of renal disease assessment. The ML (Machine Learning) approaches are crucial for illness diagnosis and early-stage diagnosis. This project’s primary goal is to evaluate the renal disease risk probability stages. It is created for classification methods that are used as meta multistage classifiers to define the danger stage. The techniques are broken up into different stages to complete the goal. The conventional data of the first module is preprocessed data. The methods used to calculate pre-processing are label encoding and standard scalar. Meta classifiers are used in extra tree classifiers to process the data along with some classifiers like K-Nearest neighbor and Random Forest. As a result, the kidney infection risk stage is known. By using meta classifiers to the Random Forest tree, a better accuracy has been obtained when compared to the existing methods.
{"title":"Detecting Renal Disease using Meta-Classifiers","authors":"Lohitha B, Adithya V, Yasaswi Aparna N, H. R., Srithar S, Aravinth S S","doi":"10.1109/IDCIoT56793.2023.10053551","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053551","url":null,"abstract":"Because of the elevated risk of illness and fatality, chronic renal disease is regarded as a serious health issue. Renal disease is also called kidney disease. Kidney infections are particularly challenging to diagnose since they progress slowly and continuously. For the same reason, a lot of patients wait until the very end stage to diagnose their condition. It’s critical to have trustworthy methods in the early stage of renal disease assessment. The ML (Machine Learning) approaches are crucial for illness diagnosis and early-stage diagnosis. This project’s primary goal is to evaluate the renal disease risk probability stages. It is created for classification methods that are used as meta multistage classifiers to define the danger stage. The techniques are broken up into different stages to complete the goal. The conventional data of the first module is preprocessed data. The methods used to calculate pre-processing are label encoding and standard scalar. Meta classifiers are used in extra tree classifiers to process the data along with some classifiers like K-Nearest neighbor and Random Forest. As a result, the kidney infection risk stage is known. By using meta classifiers to the Random Forest tree, a better accuracy has been obtained when compared to the existing methods.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"67 1","pages":"953-957"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77287129","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}
With the rapid increase in global data volume, various factors like low latency, high efficiency video surveillance is impossible to achieve in a centralized cloud computing model. Therefore, this paper proposes a distributed computing model for intelligent video surveillance system. This paper presents a smart video surveillance system which can execute Deep Learning algorithms in low power consumption embedded de vices. The proposed intelligent video surveillance system based on the edge computing consists of multi-camera for smart cities and homes. In general, the sending of original video surveillance data to the centralized computing model is too much time consuming and this will keep us far away to achieve our objective of real time data transmission so through this paper the edge computing technique is proposed, the idea is perform computation locally at the edge devices and then the computed data will be sent to the centralized computing model which is capable of performing the real time video surveillance by using the deep learning algorithm.
{"title":"An Intelligent Video Surveillance System using Edge Computing based Deep Learning Model","authors":"Rudra Pratap Singh, Harshit Srivastava, Hitesh Gautam, Rohan Shukla, Rajendra Kumar Dwivedi","doi":"10.1109/IDCIoT56793.2023.10053404","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053404","url":null,"abstract":"With the rapid increase in global data volume, various factors like low latency, high efficiency video surveillance is impossible to achieve in a centralized cloud computing model. Therefore, this paper proposes a distributed computing model for intelligent video surveillance system. This paper presents a smart video surveillance system which can execute Deep Learning algorithms in low power consumption embedded de vices. The proposed intelligent video surveillance system based on the edge computing consists of multi-camera for smart cities and homes. In general, the sending of original video surveillance data to the centralized computing model is too much time consuming and this will keep us far away to achieve our objective of real time data transmission so through this paper the edge computing technique is proposed, the idea is perform computation locally at the edge devices and then the computed data will be sent to the centralized computing model which is capable of performing the real time video surveillance by using the deep learning algorithm.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"1 1","pages":"439-444"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73149756","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 : 2023-01-05DOI: 10.1109/IDCIoT56793.2023.10053543
Nemanja Milutinovic, S. Čabarkapa, M. Zivkovic, Milos Antonijevic, Djordje Mladenovic, N. Bačanin
From 2015 to 2022, healthcare 4.0 has made revolutionary impacts on health services. It includes machine learning (ML), internet of things (IoT), fog computing and cloud computing. The utilization of machine learning approaches supplied by IoT advances employing fog and cloud computing principles improves the performance and accuracy of healthcare models. These concepts bounded together are distinguished in their application with the researchers as they dominate alongside the best results. Inspirited by the mathematical traits of sine and cosine functions, the sine cosine algorithm (SCA) generates numerous initial random candidate solutions with the goal of fluctuation outwards or towards the ideal answer. The metaheuristic algorithm can be applied for optimization of an artificial neural network (ANN) on which the Healthcare 4.0 relies. The solution has been tested on four diverse datasets in this field as well as the results of those tests have been compared to those of other hybrid solutions with the use of same datasets as the suggested solution. The results are in the favor of the novel method, as it obtains general advantage over all tests.
{"title":"Tuning Artificial Neural Network for Healthcare 4.0. by Sine Cosine Algorithm","authors":"Nemanja Milutinovic, S. Čabarkapa, M. Zivkovic, Milos Antonijevic, Djordje Mladenovic, N. Bačanin","doi":"10.1109/IDCIoT56793.2023.10053543","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053543","url":null,"abstract":"From 2015 to 2022, healthcare 4.0 has made revolutionary impacts on health services. It includes machine learning (ML), internet of things (IoT), fog computing and cloud computing. The utilization of machine learning approaches supplied by IoT advances employing fog and cloud computing principles improves the performance and accuracy of healthcare models. These concepts bounded together are distinguished in their application with the researchers as they dominate alongside the best results. Inspirited by the mathematical traits of sine and cosine functions, the sine cosine algorithm (SCA) generates numerous initial random candidate solutions with the goal of fluctuation outwards or towards the ideal answer. The metaheuristic algorithm can be applied for optimization of an artificial neural network (ANN) on which the Healthcare 4.0 relies. The solution has been tested on four diverse datasets in this field as well as the results of those tests have been compared to those of other hybrid solutions with the use of same datasets as the suggested solution. The results are in the favor of the novel method, as it obtains general advantage over all tests.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"29 1","pages":"510-513"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76088499","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 : 2023-01-05DOI: 10.1109/IDCIoT56793.2023.10053428
Barsha Biswas, R. Yadav
Around 60.3% of land in India is used for agricultural purposes and the whole population depends on agriculture. That’s why crop yield is very crucial to get high agricultural output. The economical loss will be very high if the agricultural output is low. So, that’s why the diagnosis of disease in plants is very important. And the detection should be in the early stage not in a later stage. Using Deep Learning (DL) i.e. a branch of Artificial Intelligence (AI), a farmer can detect plant diseases very easily. In Deep Learning(DL), Convolutional Neural Networks (CNNs) are a cutting-edge method for image classification tasks. And Plant Disease Detection is an image classification task in which image is given as input and a class of plant disease is obtained as an output. This research study reviews the CNN-based approaches that are used to detect various diseases in plants.
{"title":"A Review of Convolutional Neural Network-based Approaches for Disease Detection in Plants","authors":"Barsha Biswas, R. Yadav","doi":"10.1109/IDCIoT56793.2023.10053428","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053428","url":null,"abstract":"Around 60.3% of land in India is used for agricultural purposes and the whole population depends on agriculture. That’s why crop yield is very crucial to get high agricultural output. The economical loss will be very high if the agricultural output is low. So, that’s why the diagnosis of disease in plants is very important. And the detection should be in the early stage not in a later stage. Using Deep Learning (DL) i.e. a branch of Artificial Intelligence (AI), a farmer can detect plant diseases very easily. In Deep Learning(DL), Convolutional Neural Networks (CNNs) are a cutting-edge method for image classification tasks. And Plant Disease Detection is an image classification task in which image is given as input and a class of plant disease is obtained as an output. This research study reviews the CNN-based approaches that are used to detect various diseases in plants.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"111 1","pages":"514-518"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80582738","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 : 2023-01-05DOI: 10.1109/IDCIoT56793.2023.10053533
P. Purohit, Awanit Kumar, S. Degadwala
In quantum-based computing, the communication uses quantum bits rather than digital bits to address the information. This work proposes to employ encrypted communication for sharing the keys called bits in the Quantum framework. Cloud computing is one of the services, which supports both communication and quantum registering. For small to huge computing tasks, providing any software, platform, or infrastructure as a service, pay-n-use technique should be possible in distributed computing. Based on the duration and amount of administration utilization, the payment will be determined. The principle aim of the distributed computing is to create a resource where anybody can get to any help, whatsoever from anyplace. Protection and security are the fundamental dangers and issues of distributed computing. To overcome such problems in distributed computing, multi-tenure and framework sub-contracting method is used. The proposed encoding technique is far speedier and more effective than traditional encryption, for example, DNA-based encryption computing. To protect the information from intruders, better safety efforts are required. This theoretical investigation recommends such a strategy to defend against certain attacks.
{"title":"Design and Development of Protected Services in Cloud Computing Environment","authors":"P. Purohit, Awanit Kumar, S. Degadwala","doi":"10.1109/IDCIoT56793.2023.10053533","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053533","url":null,"abstract":"In quantum-based computing, the communication uses quantum bits rather than digital bits to address the information. This work proposes to employ encrypted communication for sharing the keys called bits in the Quantum framework. Cloud computing is one of the services, which supports both communication and quantum registering. For small to huge computing tasks, providing any software, platform, or infrastructure as a service, pay-n-use technique should be possible in distributed computing. Based on the duration and amount of administration utilization, the payment will be determined. The principle aim of the distributed computing is to create a resource where anybody can get to any help, whatsoever from anyplace. Protection and security are the fundamental dangers and issues of distributed computing. To overcome such problems in distributed computing, multi-tenure and framework sub-contracting method is used. The proposed encoding technique is far speedier and more effective than traditional encryption, for example, DNA-based encryption computing. To protect the information from intruders, better safety efforts are required. This theoretical investigation recommends such a strategy to defend against certain attacks.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"106 1","pages":"985-988"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80743738","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 : 2023-01-05DOI: 10.1109/IDCIoT56793.2023.10053491
S. Pandey, Vanshika, Anshul, Rajendra Kumar Dwivedi
Health is one of the most important aspects of human life. For a healthy society, the healthcare sector must be reliable and efficient as much as possible. Blockchain networks can be used to provide innovative solutions to store and share patient data among hospitals, pharmaceutical companies, diagnostic labs, and physicians. Blockchain applications are capable of building trust and detecting fatal errors in the sharing of data. So blockchain can help in the security, effectiveness, and transparency of sharing medical data in the healthcare industry. Using this technology, medical institutions can get quality and trustworthy knowledge which they can use to improve the analysis of patient data and help in the good diagnosis of health issues. A thorough study and analysis have been done to explore the opportunities that blockchain technology can provide to improve the healthcare industry. The various characteristics, methods, and smooth workflow processes of blockchain technology are discussed in diagrams as a potential means of improving global healthcare. The article concludes by listing and analyzing various significant blockchain applications for the healthcare industry. It can help in implementing the smart contract, nonce concepts, and P2P distributed ledgers which ensure the immutability of the shared data. The methods discussed in this paper improve the security of Electronic Health Records (EHRs). The data of patients are taken directly via IoT devices which are used as medical instruments for patients which increases the reliability of data as there is no human interference in between hence human error is eliminated. The proposed technique implementation results show that the security of data has increased and it is more efficient.
{"title":"A Secure Design of Healthcare System with Blockchain and Internet of Things (IoT)","authors":"S. Pandey, Vanshika, Anshul, Rajendra Kumar Dwivedi","doi":"10.1109/IDCIoT56793.2023.10053491","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053491","url":null,"abstract":"Health is one of the most important aspects of human life. For a healthy society, the healthcare sector must be reliable and efficient as much as possible. Blockchain networks can be used to provide innovative solutions to store and share patient data among hospitals, pharmaceutical companies, diagnostic labs, and physicians. Blockchain applications are capable of building trust and detecting fatal errors in the sharing of data. So blockchain can help in the security, effectiveness, and transparency of sharing medical data in the healthcare industry. Using this technology, medical institutions can get quality and trustworthy knowledge which they can use to improve the analysis of patient data and help in the good diagnosis of health issues. A thorough study and analysis have been done to explore the opportunities that blockchain technology can provide to improve the healthcare industry. The various characteristics, methods, and smooth workflow processes of blockchain technology are discussed in diagrams as a potential means of improving global healthcare. The article concludes by listing and analyzing various significant blockchain applications for the healthcare industry. It can help in implementing the smart contract, nonce concepts, and P2P distributed ledgers which ensure the immutability of the shared data. The methods discussed in this paper improve the security of Electronic Health Records (EHRs). The data of patients are taken directly via IoT devices which are used as medical instruments for patients which increases the reliability of data as there is no human interference in between hence human error is eliminated. The proposed technique implementation results show that the security of data has increased and it is more efficient.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"8 1","pages":"105-111"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87783333","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 : 2023-01-05DOI: 10.1109/IDCIoT56793.2023.10053462
Durgesh Kumar, Rajendra Kumar Dwivedi
In the present situation, most people are not satisfied with the final result of the voting system. This is because the current system for voting is centralized and fully controlled by the election commission. So, there is a chance that the central body can be compromised or hacked and the final result can be tempered. In this direction, a decentralized, Blockchain and IoT based methodology for voting system is devised and presented in this paper. Blockchain is totally transparent, secured and immutable technique because it uses concept like encryption, decryption, hash function, consensus and Merkle tree etc. which make Blockchain Technology an appropriate platform for storing and sharing the data in a secured and anonymous manner. IoT makes use of biometric sensors using which people can cast their votes in not only physical mode but also in digital mode. As a response, a message is received to the owner for casting his vote to ensure the authentication. In this way, the present voting system is more secure and trust-worthy by using the properties of both Blockchain and IoT, and therefore, election process in the democratic countries is valued more. The proposed method ensures security as well as reduces the computational time as compared to the existing approaches.
{"title":"Blockchain and Internet of Things (IoT) Enabled Smart E-Voting System","authors":"Durgesh Kumar, Rajendra Kumar Dwivedi","doi":"10.1109/IDCIoT56793.2023.10053462","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053462","url":null,"abstract":"In the present situation, most people are not satisfied with the final result of the voting system. This is because the current system for voting is centralized and fully controlled by the election commission. So, there is a chance that the central body can be compromised or hacked and the final result can be tempered. In this direction, a decentralized, Blockchain and IoT based methodology for voting system is devised and presented in this paper. Blockchain is totally transparent, secured and immutable technique because it uses concept like encryption, decryption, hash function, consensus and Merkle tree etc. which make Blockchain Technology an appropriate platform for storing and sharing the data in a secured and anonymous manner. IoT makes use of biometric sensors using which people can cast their votes in not only physical mode but also in digital mode. As a response, a message is received to the owner for casting his vote to ensure the authentication. In this way, the present voting system is more secure and trust-worthy by using the properties of both Blockchain and IoT, and therefore, election process in the democratic countries is valued more. The proposed method ensures security as well as reduces the computational time as compared to the existing approaches.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"5 1","pages":"28-34"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85503016","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 : 2023-01-05DOI: 10.1109/IDCIoT56793.2023.10053545
Shanmukha Kantimahanthi, J. Prasad, Sravan Chanamolu, Kavyasree Kommaraju
Cyber-physical systems (CPS) enabled by the Internet of Things (IoT) provide unique security challenges since solutions designed for traditional Operational Technology (OT) and Information Technology (IT) systems may not be adequate in a Cyber-Physical System environment. With that in mind, this research introduces a two-tiered integrated attack detection and attack attribution framework ideal for Cyber-physical systems (CPS), and more particularly in an Industrial Control System (ICS). In order to identify assaults in unbalanced ICS settings, in the first phase, a unique ensemble deep-representational learning model is coupled with a decision tree classifier. In the next phase, an attack attribution ensemble deep neural network is developed. Datasets from the MODBUS and the natural gas pipeline industry are used to test the accuracy of the proposed model. The proposed model outperforms comparable models with a similar degree of computational complexity.
{"title":"Machine Learning Approaches in Cyber Attack Detection and Characterization in IoT enabled Cyber-Physical Systems","authors":"Shanmukha Kantimahanthi, J. Prasad, Sravan Chanamolu, Kavyasree Kommaraju","doi":"10.1109/IDCIoT56793.2023.10053545","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053545","url":null,"abstract":"Cyber-physical systems (CPS) enabled by the Internet of Things (IoT) provide unique security challenges since solutions designed for traditional Operational Technology (OT) and Information Technology (IT) systems may not be adequate in a Cyber-Physical System environment. With that in mind, this research introduces a two-tiered integrated attack detection and attack attribution framework ideal for Cyber-physical systems (CPS), and more particularly in an Industrial Control System (ICS). In order to identify assaults in unbalanced ICS settings, in the first phase, a unique ensemble deep-representational learning model is coupled with a decision tree classifier. In the next phase, an attack attribution ensemble deep neural network is developed. Datasets from the MODBUS and the natural gas pipeline industry are used to test the accuracy of the proposed model. The proposed model outperforms comparable models with a similar degree of computational complexity.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"33 1","pages":"136-142"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91146242","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 : 2023-01-05DOI: 10.1109/IDCIoT56793.2023.10053470
Deemah Alosail, Hussa Aldolah, Layla Alabdulwahab, A. Bashar, Majid Khan
The deaf community in our society has a right to live a comfortable and respectable life by having communication with normal people without any hurdles or impediments. To address this objective, several research attempts have been made to develop smart gloves to provide a means of converting sign language to speech or text. This research work has attempted to design, implement and test non-visual-based smart glove to improve performance accuracy and reduce implementation complexity. More specifically, five flex sensors and an accelerometer are used to enable sign language recognition and its further conversion into speech and textual information. Further, the prominent Machine Learning (ML) classifiers (LR, SVM, MLP and RF) are used for recognising both American Sign Language (ASL) and Arabic Sign Language (ArSL). Finally, a classification accuracy of 99.7% for ASL and 99.8% for ArSL with Random Forests (RF) classifier has been achieved. By considering the Feature Importance, the accelerometer features are considered as dominant features in recognizing the sign language when compared to the flex sensor features. In order to further advance this research work, the implementation and performance aspects of non-vision and vision-based sign language recognition can be compared.
{"title":"Smart Glove for Bi-lingual Sign Language Recognition using Machine Learning","authors":"Deemah Alosail, Hussa Aldolah, Layla Alabdulwahab, A. Bashar, Majid Khan","doi":"10.1109/IDCIoT56793.2023.10053470","DOIUrl":"https://doi.org/10.1109/IDCIoT56793.2023.10053470","url":null,"abstract":"The deaf community in our society has a right to live a comfortable and respectable life by having communication with normal people without any hurdles or impediments. To address this objective, several research attempts have been made to develop smart gloves to provide a means of converting sign language to speech or text. This research work has attempted to design, implement and test non-visual-based smart glove to improve performance accuracy and reduce implementation complexity. More specifically, five flex sensors and an accelerometer are used to enable sign language recognition and its further conversion into speech and textual information. Further, the prominent Machine Learning (ML) classifiers (LR, SVM, MLP and RF) are used for recognising both American Sign Language (ASL) and Arabic Sign Language (ArSL). Finally, a classification accuracy of 99.7% for ASL and 99.8% for ArSL with Random Forests (RF) classifier has been achieved. By considering the Feature Importance, the accelerometer features are considered as dominant features in recognizing the sign language when compared to the flex sensor features. In order to further advance this research work, the implementation and performance aspects of non-vision and vision-based sign language recognition can be compared.","PeriodicalId":60583,"journal":{"name":"物联网技术","volume":"29 2 1","pages":"409-415"},"PeriodicalIF":0.0,"publicationDate":"2023-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81005796","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：