Pub Date : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454247
Haymontee Khan, Faria Soroni, Syed Jafar Sadek Mahmood, Noel Mannan, Mohammad Monirujjaman Khan
This paper presents an innovative application for students to study and understand their coursework without any external help from a private tutor. The system uses Augmented Reality (AR) to provide hands on experience for the students. The presented system also supports Virtual Reality (VR) that enriches this process and immerses the users into a fun and productive learning experience. Moreover, the system introduces an industry first Artificial intelligence (AI) based study guide that directs students towards necessary topics and advises them on what to improve on. All the core system features are implemented and are accessible via two mediums. First, a standalone mobile phone application. Second, a dedicated web portal.
{"title":"Education System for Bangladesh Using Augmented Reality, Virtual Reality and Artificial Intelligence","authors":"Haymontee Khan, Faria Soroni, Syed Jafar Sadek Mahmood, Noel Mannan, Mohammad Monirujjaman Khan","doi":"10.1109/AIIoT52608.2021.9454247","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454247","url":null,"abstract":"This paper presents an innovative application for students to study and understand their coursework without any external help from a private tutor. The system uses Augmented Reality (AR) to provide hands on experience for the students. The presented system also supports Virtual Reality (VR) that enriches this process and immerses the users into a fun and productive learning experience. Moreover, the system introduces an industry first Artificial intelligence (AI) based study guide that directs students towards necessary topics and advises them on what to improve on. All the core system features are implemented and are accessible via two mediums. First, a standalone mobile phone application. Second, a dedicated web portal.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124354065","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454237
Panagiotis Savvidis, G. Papakostas
The current work in this paper inspired by the concepts of Edge Computing, Machine Learning, Computer Vision and Internet of Things (IoT). This synergy is used for monitoring apple orchard yield and more specific the detection and information extraction for apple harvesting purposes in the agriculture field. The above concept utilizes the means for a low power information relay using LoRaWAN (Low Power Wide Area Network) protocol designed to connect battery operated “things” with the internet in regional or global topology. Image acquisition and data are processed on a battery driven edge device away from the grid and on site. The proposition implementing a full YoloV4 framework in a single board computer (SBC) equipped with a proper camera and by using custom-trained weights seems to be a feasible solution. The performance of the proposed approach for good apple detection is up to 66.89% for complex dense environments. These preliminary results reveal the feasibility of this edge computing approach utilizing Artificial Intelligence and IoT technologies.
{"title":"Remote Crop Sensing with IoT and AI on the Edge","authors":"Panagiotis Savvidis, G. Papakostas","doi":"10.1109/AIIoT52608.2021.9454237","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454237","url":null,"abstract":"The current work in this paper inspired by the concepts of Edge Computing, Machine Learning, Computer Vision and Internet of Things (IoT). This synergy is used for monitoring apple orchard yield and more specific the detection and information extraction for apple harvesting purposes in the agriculture field. The above concept utilizes the means for a low power information relay using LoRaWAN (Low Power Wide Area Network) protocol designed to connect battery operated “things” with the internet in regional or global topology. Image acquisition and data are processed on a battery driven edge device away from the grid and on site. The proposition implementing a full YoloV4 framework in a single board computer (SBC) equipped with a proper camera and by using custom-trained weights seems to be a feasible solution. The performance of the proposed approach for good apple detection is up to 66.89% for complex dense environments. These preliminary results reveal the feasibility of this edge computing approach utilizing Artificial Intelligence and IoT technologies.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130053271","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454216
Hagar Louye Elghazy, M. Fakhr
Computer segmentation of abdominal organs using CT and MRI images can benefit diagnosis, treatment, and workload management. In recent years, UNETs have been widely used in medical image segmentation for their precise accuracy. Most of the UNETs current solutions rely on the use of single data modality. Recently, it has been shown that learning from more than one modality at a time can significantly enhance the segmentation accuracy, however most of available multi-modal datasets are not large enough for training complex architectures. In this paper, we worked on a small dataset and proposed a multi-modal dual-stream UNET architecture that learns from unpaired MRI and CT image modalities to improve the segmentation accuracy on each individual one. We tested the practicality of the proposed architecture on Task 1 of the CHAOS segmentation challenge. Results showed that multi-modal/multi-stream learning improved accuracy over single modality learning and that using UNET in the dual stream was superior than using a standard FCN. A “Dice” score of 96.78 was achieved on CT images. To the best of our knowledge, this is one of the highest reported scores yet.
{"title":"Multi-Modal Multi-Stream UNET Model for Liver Segmentation","authors":"Hagar Louye Elghazy, M. Fakhr","doi":"10.1109/AIIoT52608.2021.9454216","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454216","url":null,"abstract":"Computer segmentation of abdominal organs using CT and MRI images can benefit diagnosis, treatment, and workload management. In recent years, UNETs have been widely used in medical image segmentation for their precise accuracy. Most of the UNETs current solutions rely on the use of single data modality. Recently, it has been shown that learning from more than one modality at a time can significantly enhance the segmentation accuracy, however most of available multi-modal datasets are not large enough for training complex architectures. In this paper, we worked on a small dataset and proposed a multi-modal dual-stream UNET architecture that learns from unpaired MRI and CT image modalities to improve the segmentation accuracy on each individual one. We tested the practicality of the proposed architecture on Task 1 of the CHAOS segmentation challenge. Results showed that multi-modal/multi-stream learning improved accuracy over single modality learning and that using UNET in the dual stream was superior than using a standard FCN. A “Dice” score of 96.78 was achieved on CT images. To the best of our knowledge, this is one of the highest reported scores yet.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130666763","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454228
MinhTri Tran, A. Kuwana, Haruo Kobayashi
This paper presents several proposed motion models of high-order physical systems in three main concepts called macro-scale, regular-scale, and nano-scale. In fact, it is very difficult to find an exact numerical solution for the high-order differential equations because all numerical methods only yield the approximate solutions. In addition, loop gain is not widely used in many negative feedback systems because it is an approximation value. To overcome the limitations of the high-order differential equations and the loop gain, the waveforms of the physical periodic motions are expressed by helix functions at time variation, and the characteristics of complex functions are used to examine the behaviors of the transmission spaces and the transmission networks in the different motion models including the Earth's motion, the simple pendulum systems, and the electronic systems. Furthermore, the force of attraction and the friction or the resistance in the different scales obey the conservation law and the superposition principle; therefore, three superposition formulas are introduced to derive the transfer functions in high-order mechatronic systems. The operating regions, the effects of the overshoot phenomena, the breaking chemical bonds, and the difference between negative and positive feedbacks in these systems are also introduced. As a result, the use of complex functions, helix waves, and superposition principle leads to a complete control theory with which many behaviors of the physical systems can be explained and predicted easily.
{"title":"Study of Behaviors of Motion Models in High-Order Systems","authors":"MinhTri Tran, A. Kuwana, Haruo Kobayashi","doi":"10.1109/AIIoT52608.2021.9454228","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454228","url":null,"abstract":"This paper presents several proposed motion models of high-order physical systems in three main concepts called macro-scale, regular-scale, and nano-scale. In fact, it is very difficult to find an exact numerical solution for the high-order differential equations because all numerical methods only yield the approximate solutions. In addition, loop gain is not widely used in many negative feedback systems because it is an approximation value. To overcome the limitations of the high-order differential equations and the loop gain, the waveforms of the physical periodic motions are expressed by helix functions at time variation, and the characteristics of complex functions are used to examine the behaviors of the transmission spaces and the transmission networks in the different motion models including the Earth's motion, the simple pendulum systems, and the electronic systems. Furthermore, the force of attraction and the friction or the resistance in the different scales obey the conservation law and the superposition principle; therefore, three superposition formulas are introduced to derive the transfer functions in high-order mechatronic systems. The operating regions, the effects of the overshoot phenomena, the breaking chemical bonds, and the difference between negative and positive feedbacks in these systems are also introduced. As a result, the use of complex functions, helix waves, and superposition principle leads to a complete control theory with which many behaviors of the physical systems can be explained and predicted easily.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116426903","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454210
S. Joshi, M. El-Sharkawy
Deep learning models require a lot of computation and memory, so they can only be run on high-performance computing platforms such as CPUs or GPUs. However, due to resource, energy, and real-time constraints, they often fail to meet portable requirements. As a result, there is an increasing interest in real-time object recognition solutions based on CNNs, which are typically implemented on embedded systems with limited resources and energy consumption. Recently, hardware accelerators have been developed to provide the computing power needed by AI and machine learning tools. These edge accelerators deliver high-performance hardware while maintaining the needed accuracy for the task at hand. This paper takes a step forward by suggesting a design approach for porting CNNs to low-resource embedded systems, bridging the gap between deep learning models and embedded edge systems. To complete our task, we employ closer computing approaches to minimize the computational load and memory consumption of the computer while maintaining impressive deployment performance. HBONext is one of those models that was designed to be easily deployable on embedded and mobile devices. We demonstrate how to use NXP BlueBox 2.0 to introduce a real-time HBONext image classifier in this work. Incorporating this concept into this hardware has been a huge success due to its limited architectural scale of 3 MB. This model was trained and validated using the CIFAR10 data set, which performed exceptionally well due to its smaller size and higher accuracy.
{"title":"Hardware Deployment of HBONext using NXP Bluebox 2.0","authors":"S. Joshi, M. El-Sharkawy","doi":"10.1109/AIIoT52608.2021.9454210","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454210","url":null,"abstract":"Deep learning models require a lot of computation and memory, so they can only be run on high-performance computing platforms such as CPUs or GPUs. However, due to resource, energy, and real-time constraints, they often fail to meet portable requirements. As a result, there is an increasing interest in real-time object recognition solutions based on CNNs, which are typically implemented on embedded systems with limited resources and energy consumption. Recently, hardware accelerators have been developed to provide the computing power needed by AI and machine learning tools. These edge accelerators deliver high-performance hardware while maintaining the needed accuracy for the task at hand. This paper takes a step forward by suggesting a design approach for porting CNNs to low-resource embedded systems, bridging the gap between deep learning models and embedded edge systems. To complete our task, we employ closer computing approaches to minimize the computational load and memory consumption of the computer while maintaining impressive deployment performance. HBONext is one of those models that was designed to be easily deployable on embedded and mobile devices. We demonstrate how to use NXP BlueBox 2.0 to introduce a real-time HBONext image classifier in this work. Incorporating this concept into this hardware has been a huge success due to its limited architectural scale of 3 MB. This model was trained and validated using the CIFAR10 data set, which performed exceptionally well due to its smaller size and higher accuracy.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129706789","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454177
Sudman Bin Manjur, Nahian Noshin Nur, Md. Mushfiqur Rahman, Rohimul Basunia, Mohammad Monirujjaman Khan
Proper menstrual hygiene management is vital to the poise and strength of women and young ladies. In any case, it is a disregarded issue both in the overall individuals and health sectors, prompting an emergency of information, offices and hygienic practice. To eliminate the feminine cleanliness of the board issues and social issues, we present to you a web application to raise awareness among individuals about menstruation. This is essentially intended for the youthful ages to show them evidently. We have chiefly utilized PHP 7, HTML, C# and CSS for all coding and information putting away system. Word Press, Adobe Flash and Blender are utilized to do the animations, videos, design and other kinds of things. There will be interactive animation questions according to the child's understanding. There will also be short fun quiz games, different methods of explanation for both boys and girls, options to ask questions from experts and many more things. This is also an attempt to normalize menstruation among people and to minimize taboos and misconceptions on this topic.
{"title":"Educational Web Application for Young People to Raise Awareness on Menstruation","authors":"Sudman Bin Manjur, Nahian Noshin Nur, Md. Mushfiqur Rahman, Rohimul Basunia, Mohammad Monirujjaman Khan","doi":"10.1109/AIIoT52608.2021.9454177","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454177","url":null,"abstract":"Proper menstrual hygiene management is vital to the poise and strength of women and young ladies. In any case, it is a disregarded issue both in the overall individuals and health sectors, prompting an emergency of information, offices and hygienic practice. To eliminate the feminine cleanliness of the board issues and social issues, we present to you a web application to raise awareness among individuals about menstruation. This is essentially intended for the youthful ages to show them evidently. We have chiefly utilized PHP 7, HTML, C# and CSS for all coding and information putting away system. Word Press, Adobe Flash and Blender are utilized to do the animations, videos, design and other kinds of things. There will be interactive animation questions according to the child's understanding. There will also be short fun quiz games, different methods of explanation for both boys and girls, options to ask questions from experts and many more things. This is also an attempt to normalize menstruation among people and to minimize taboos and misconceptions on this topic.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131167574","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454206
Muath A. Obaidat, Joseph Brown, Awny Alnusair
Adaptive honeypots are being widely proposed as a more powerful alternative to the traditional honeypot model. Just as with typical honeypots, however, one of the most important concerns of an adaptive honeypot is environment deception in order to make sure an adversary cannot fingerprint the honeypot. The threat of fingerprinting hints at a greater underlying concern, however; this being that honeypots are only effective because an adversary does not know that the environment on which they are operating is a honeypot. What has not been widely discussed in the context of adaptive honeypots is that they actually have an inherently increased level of susceptibility to this threat. Honeypots not only bear increased risks when an adversary knows they are a honeypot rather than a native system, but they are only effective as adaptable entities if one does not know that the honeypot environment they are operating on is adaptive as wekk. Thus, if adaptive honeypots become commonplace - or, instead, if attackers even have an inkling that an adaptive honeypot may exist on any given network, a new attack which could develop is a “blind confusion attack”; a form of connection which simply makes an assumption all environments are adaptive honeypots, and instead of attempting to perform a malicious strike on a given entity, opts to perform non-malicious behavior in specified and/or random patterns to confuse an adaptive network's learning.
{"title":"Blind Attack Flaws in Adaptive Honeypot Strategies","authors":"Muath A. Obaidat, Joseph Brown, Awny Alnusair","doi":"10.1109/AIIoT52608.2021.9454206","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454206","url":null,"abstract":"Adaptive honeypots are being widely proposed as a more powerful alternative to the traditional honeypot model. Just as with typical honeypots, however, one of the most important concerns of an adaptive honeypot is environment deception in order to make sure an adversary cannot fingerprint the honeypot. The threat of fingerprinting hints at a greater underlying concern, however; this being that honeypots are only effective because an adversary does not know that the environment on which they are operating is a honeypot. What has not been widely discussed in the context of adaptive honeypots is that they actually have an inherently increased level of susceptibility to this threat. Honeypots not only bear increased risks when an adversary knows they are a honeypot rather than a native system, but they are only effective as adaptable entities if one does not know that the honeypot environment they are operating on is adaptive as wekk. Thus, if adaptive honeypots become commonplace - or, instead, if attackers even have an inkling that an adaptive honeypot may exist on any given network, a new attack which could develop is a “blind confusion attack”; a form of connection which simply makes an assumption all environments are adaptive honeypots, and instead of attempting to perform a malicious strike on a given entity, opts to perform non-malicious behavior in specified and/or random patterns to confuse an adaptive network's learning.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127750161","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454181
Ilan Yehuda Granot, C. Wu, Z. Or-Bach
We propose a method for a recommender system for generating web-navigation suggestions. The purpose of this system is to assist its users by providing them suggestions for possible desired next steps whenever they get stuck in using any software. We are able to achieve this goal by leveraging the principal of “crowd-sourcing”. Specifically, we leverage the crowd's knowledge under the assumption that there are cohesive groups of experienced and novice users. Therefore, we present an algorithm that measures the right heuristics in order to classify users by their experience, and then relates these users with association rules of web-navigation derived from frequent patterns mining. In this paper we introduce our method, compare it with other current solutions in the field, outline the proposed algorithm, and present an experiment which serves as our proof-of-concept.
{"title":"Improving the Relevance of a Web Navigation Recommender System Using Categorization of Users' Experience","authors":"Ilan Yehuda Granot, C. Wu, Z. Or-Bach","doi":"10.1109/AIIoT52608.2021.9454181","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454181","url":null,"abstract":"We propose a method for a recommender system for generating web-navigation suggestions. The purpose of this system is to assist its users by providing them suggestions for possible desired next steps whenever they get stuck in using any software. We are able to achieve this goal by leveraging the principal of “crowd-sourcing”. Specifically, we leverage the crowd's knowledge under the assumption that there are cohesive groups of experienced and novice users. Therefore, we present an algorithm that measures the right heuristics in order to classify users by their experience, and then relates these users with association rules of web-navigation derived from frequent patterns mining. In this paper we introduce our method, compare it with other current solutions in the field, outline the proposed algorithm, and present an experiment which serves as our proof-of-concept.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128549559","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 : 2021-05-10DOI: 10.1109/AIIoT52608.2021.9454171
P. Subashini, P. V. H. Kumar, S. Lekshmi, M. Krishnaveni, T. Dhivyaprabha
Sea surface is rough when the weather condition at sea is rough due to strong wind, waves, swell and storms. Under the rough sea condition, the propagation of radar energy and the subsequent radar coverage is strongly influenced by various atmospheric effects, such as, strong wind, wave height, weather condition, oceanic currents and rainstorms. The identification of ship wakes in Synthetic Aperture Radar (SAR) image under the rough sea condition is viewed as a highly complex task for the real time monitoring and surveillance applications. It becomes a quite big challenge due to coherent radiation of backscattering signals and the multiplicative speckle noise found in SAR images. The objective of this work is to develop an optimized Discrete Wavelet Transform (DWT) based on Synergistic Fibroblast Optimization (SFO) algorithm for filtering speckle noise in SAR image which are captured under rough sea condition. An improved filtering technique is tested with the real time SAR images acquired from European Space Agency (ESA) sentinel scientific data hub and its efficacy is further validated by employing Discrete Radon Transform (DRT) method to detect ship wakes (linear signature) in SAR image under rough sea surface. The performance of SFO based wavelet transform is evaluated and compared with conventional DWT families, namely, daubechies, coiflet, symlet, discrete meyer, biorthogonal and reverse biorthogonal to conduct the better investigation of this study. Investigation of results illustrates the effectiveness of optimized method, in terms of, noise suppression and its implication on radon transform method to localize the identification of ship wakes in SAR imagery.
{"title":"Improved Noise Filtering Technique For Wake Detection In SAR Image Under Rough Sea Condition","authors":"P. Subashini, P. V. H. Kumar, S. Lekshmi, M. Krishnaveni, T. Dhivyaprabha","doi":"10.1109/AIIoT52608.2021.9454171","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454171","url":null,"abstract":"Sea surface is rough when the weather condition at sea is rough due to strong wind, waves, swell and storms. Under the rough sea condition, the propagation of radar energy and the subsequent radar coverage is strongly influenced by various atmospheric effects, such as, strong wind, wave height, weather condition, oceanic currents and rainstorms. The identification of ship wakes in Synthetic Aperture Radar (SAR) image under the rough sea condition is viewed as a highly complex task for the real time monitoring and surveillance applications. It becomes a quite big challenge due to coherent radiation of backscattering signals and the multiplicative speckle noise found in SAR images. The objective of this work is to develop an optimized Discrete Wavelet Transform (DWT) based on Synergistic Fibroblast Optimization (SFO) algorithm for filtering speckle noise in SAR image which are captured under rough sea condition. An improved filtering technique is tested with the real time SAR images acquired from European Space Agency (ESA) sentinel scientific data hub and its efficacy is further validated by employing Discrete Radon Transform (DRT) method to detect ship wakes (linear signature) in SAR image under rough sea surface. The performance of SFO based wavelet transform is evaluated and compared with conventional DWT families, namely, daubechies, coiflet, symlet, discrete meyer, biorthogonal and reverse biorthogonal to conduct the better investigation of this study. Investigation of results illustrates the effectiveness of optimized method, in terms of, noise suppression and its implication on radon transform method to localize the identification of ship wakes in SAR imagery.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115006537","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 : 2021-03-20DOI: 10.1109/AIIoT52608.2021.9454173
Gerasimos G. Samatas, Seraphim S. Moumgiakmas, G. Papakostas
This paper highlights the trends in the field of predictive maintenance with the use of machine learning. With the continuous development of the Fourth Industrial Revolution, through IoT, the technologies that use artificial intelligence are evolving. As a result, industries have been using these technologies to optimize their production. Through scientific research conducted for this paper, conclusions were drawn about the trends in Predictive Maintenance applications with the use of machine learning bridging Artificial Intelligence and IoT. These trends are related to the types of industries in which Predictive Maintenance was applied, the models of artificial intelligence were implemented, mainly of machine learning and the types of sensors that are applied through the IoT to the applications. Six sectors were presented and the production sector was dominant as it accounted for 54.55% of total publications. In terms of artificial intelligence models, the most prevalent among ten were the Artificial Neural Networks, Support Vector Machine and Random Forest with 28.95%, 18.42% and 14.47% respectively. Finally, 12 categories of sensors emerged, of which the most widely used were the sensors of temperature and vibration with percentages of 60.71% and 46.42% correspondingly.
{"title":"Predictive Maintenance - Bridging Artificial Intelligence and IoT","authors":"Gerasimos G. Samatas, Seraphim S. Moumgiakmas, G. Papakostas","doi":"10.1109/AIIoT52608.2021.9454173","DOIUrl":"https://doi.org/10.1109/AIIoT52608.2021.9454173","url":null,"abstract":"This paper highlights the trends in the field of predictive maintenance with the use of machine learning. With the continuous development of the Fourth Industrial Revolution, through IoT, the technologies that use artificial intelligence are evolving. As a result, industries have been using these technologies to optimize their production. Through scientific research conducted for this paper, conclusions were drawn about the trends in Predictive Maintenance applications with the use of machine learning bridging Artificial Intelligence and IoT. These trends are related to the types of industries in which Predictive Maintenance was applied, the models of artificial intelligence were implemented, mainly of machine learning and the types of sensors that are applied through the IoT to the applications. Six sectors were presented and the production sector was dominant as it accounted for 54.55% of total publications. In terms of artificial intelligence models, the most prevalent among ten were the Artificial Neural Networks, Support Vector Machine and Random Forest with 28.95%, 18.42% and 14.47% respectively. Finally, 12 categories of sensors emerged, of which the most widely used were the sensors of temperature and vibration with percentages of 60.71% and 46.42% correspondingly.","PeriodicalId":443405,"journal":{"name":"2021 IEEE World AI IoT Congress (AIIoT)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120959077","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：