Pub Date : 2018-11-01DOI: 10.1109/AIEEE.2018.8592033
Eldar abanoviè, Gediminas Stankevièius, Dalius Matuzevièius
Feature description is an important step in image registration work flow. Discriminative power of feature descriptors affects feature matching performance and overall results of image registration. Deep Neural Network-based (DNN) feature descriptors are emerging trend in image registration tasks, often performing equally or better than hand-crafted ones. However, there are no learned local feature descriptors, specifically trained for human retinal image registration. In this paper we propose DNN-based feature descriptor that was trained on retinal image patches and compare it to well-known hand-crafted feature descriptors. Training dataset of image patches was compiled from nine online datasets of eye fundus images. Learned feature descriptor was compared to other descriptors using Fundus Image Registration dataset (FIRE), measuring amount of correctly matched ground truth points (Rank-1 metric) after feature description. We compare the performance of various feature descriptors applied for retinal image feature matching.
{"title":"Deep Neural Network-based Feature Descriptor for Retinal Image Registration","authors":"Eldar abanoviè, Gediminas Stankevièius, Dalius Matuzevièius","doi":"10.1109/AIEEE.2018.8592033","DOIUrl":"https://doi.org/10.1109/AIEEE.2018.8592033","url":null,"abstract":"Feature description is an important step in image registration work flow. Discriminative power of feature descriptors affects feature matching performance and overall results of image registration. Deep Neural Network-based (DNN) feature descriptors are emerging trend in image registration tasks, often performing equally or better than hand-crafted ones. However, there are no learned local feature descriptors, specifically trained for human retinal image registration. In this paper we propose DNN-based feature descriptor that was trained on retinal image patches and compare it to well-known hand-crafted feature descriptors. Training dataset of image patches was compiled from nine online datasets of eye fundus images. Learned feature descriptor was compared to other descriptors using Fundus Image Registration dataset (FIRE), measuring amount of correctly matched ground truth points (Rank-1 metric) after feature description. We compare the performance of various feature descriptors applied for retinal image feature matching.","PeriodicalId":198244,"journal":{"name":"2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114157678","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 : 2018-11-01DOI: 10.1109/AIEEE.2018.8592151
Sandis Dejus, A. Zabasta, R. Brūzgienė, L. Narbutaite, N. Kuņicina
Stable and reliable operation of the drinking water distribution system is very important for vital human needs. Moreover, its integration with the use of Wireless Sensor Networks gives a valid way for better monitoring of water quality. However, this doesn’t provide a higher resilience of the system in a case of many possible disruptions, caused by natural, inattentive or malicious factors. More or less, the vulnerability of this system performance highly depends not only on the power supply but on the data transmission quality factors, which affect a sensitive data in all stages of its delivery. Due to this, the authors proposed a novel idea by developing a digital model for monitoring of quality-focused parameters in a data delivery over Wireless Sensor Network based drinking water distribution system in Latvia. We proposed a concept of Quality of Resilience for overall system of Wireless Sensor Network for control of water distribution system. We also proposed and implemented a digital model, investigations, insights for self-organization in Wireless Sensor Networkbased water distribution system, which would help to increase the system’s resilience to various short-time disruptions. This model implements a quality control in data delivery and provides the opportunity to re-organize topology of the sensors, aiming to maintain a stable operation of a system and reduce a percentage of messages from data processing system (broker) for data retransmission in a failure of any gateway – controller. This integrated solution for self-organization of sensors and data delivery with a selected Quality of Service level would increase both - the overall system’s resilience in a case of shortterm disruptions and the stable system performance at that time.
{"title":"Quality-Focused Data Delivery in Wireless Sensor Network for Drinking Water Distribution System","authors":"Sandis Dejus, A. Zabasta, R. Brūzgienė, L. Narbutaite, N. Kuņicina","doi":"10.1109/AIEEE.2018.8592151","DOIUrl":"https://doi.org/10.1109/AIEEE.2018.8592151","url":null,"abstract":"Stable and reliable operation of the drinking water distribution system is very important for vital human needs. Moreover, its integration with the use of Wireless Sensor Networks gives a valid way for better monitoring of water quality. However, this doesn’t provide a higher resilience of the system in a case of many possible disruptions, caused by natural, inattentive or malicious factors. More or less, the vulnerability of this system performance highly depends not only on the power supply but on the data transmission quality factors, which affect a sensitive data in all stages of its delivery. Due to this, the authors proposed a novel idea by developing a digital model for monitoring of quality-focused parameters in a data delivery over Wireless Sensor Network based drinking water distribution system in Latvia. We proposed a concept of Quality of Resilience for overall system of Wireless Sensor Network for control of water distribution system. We also proposed and implemented a digital model, investigations, insights for self-organization in Wireless Sensor Networkbased water distribution system, which would help to increase the system’s resilience to various short-time disruptions. This model implements a quality control in data delivery and provides the opportunity to re-organize topology of the sensors, aiming to maintain a stable operation of a system and reduce a percentage of messages from data processing system (broker) for data retransmission in a failure of any gateway – controller. This integrated solution for self-organization of sensors and data delivery with a selected Quality of Service level would increase both - the overall system’s resilience in a case of shortterm disruptions and the stable system performance at that time.","PeriodicalId":198244,"journal":{"name":"2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125254678","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 : 2018-11-01DOI: 10.1109/AIEEE.2018.8592253
Jūlija Golosova, A. Romānovs
The Blockchain is the newest and perspective technology in modern economy. This technology can help to solve different kind of problems in the industrial sphere, such as trust, transparency, security and reliability of data processing. In theory, the use of Blockchain technology shows great and positive results, but what can say about practice? In this paper the description of the Blockchain technology, and it advantages and disadvantages are analyzed. Many already implemented applications of Blockchain technology were studied, as well as affected success or problems factors during the implementations. This paper aim is to analyze conveniences and difficulties, related to the Blockchain integration and implementation in the different fields of modern industry.
{"title":"The Advantages and Disadvantages of the Blockchain Technology","authors":"Jūlija Golosova, A. Romānovs","doi":"10.1109/AIEEE.2018.8592253","DOIUrl":"https://doi.org/10.1109/AIEEE.2018.8592253","url":null,"abstract":"The Blockchain is the newest and perspective technology in modern economy. This technology can help to solve different kind of problems in the industrial sphere, such as trust, transparency, security and reliability of data processing. In theory, the use of Blockchain technology shows great and positive results, but what can say about practice? In this paper the description of the Blockchain technology, and it advantages and disadvantages are analyzed. Many already implemented applications of Blockchain technology were studied, as well as affected success or problems factors during the implementations. This paper aim is to analyze conveniences and difficulties, related to the Blockchain integration and implementation in the different fields of modern industry.","PeriodicalId":198244,"journal":{"name":"2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129634226","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 : 2018-11-01DOI: 10.1109/AIEEE.2018.8592203
Andis Āriņš
Border gateway protocol (BGP) version 4 is routing current Internet for more than 20 years. BGP is the only protocol that has proven its routing capability of such size network with continually growing nature. Although BGP is scalable and network layer reachability information it lacks quality of service metrics like latency. BGP routing decision algorithm uses as-path length between autonomous systems (AS) as main factor for most of its best-path calculations. Because of more affordable peering architectures as-path length is generally decreasing in worldwide internet network resulting in less efficient routes for real-time internet protocol traffic. This study proposes future internet architecture consisting of multiple AS which provide latency aware routing in software defined network (SDN). For highest reliability there are bilateral latency measurements between each pair of connected autonomous systems. Latency measurement results are stored in distributed decentralized blockchain network where each AS holds its share. Each AS hosts an SDN instance for interconnection with neighbor SDN controllers over AS border. SDN instances processes latency measurement data from blockchain network, validates data integrity and manages latency aware routing for real-time data flows.
{"title":"Blockchain based Inter-domain Latency Aware Routing Proposal in Software Defined Network","authors":"Andis Āriņš","doi":"10.1109/AIEEE.2018.8592203","DOIUrl":"https://doi.org/10.1109/AIEEE.2018.8592203","url":null,"abstract":"Border gateway protocol (BGP) version 4 is routing current Internet for more than 20 years. BGP is the only protocol that has proven its routing capability of such size network with continually growing nature. Although BGP is scalable and network layer reachability information it lacks quality of service metrics like latency. BGP routing decision algorithm uses as-path length between autonomous systems (AS) as main factor for most of its best-path calculations. Because of more affordable peering architectures as-path length is generally decreasing in worldwide internet network resulting in less efficient routes for real-time internet protocol traffic. This study proposes future internet architecture consisting of multiple AS which provide latency aware routing in software defined network (SDN). For highest reliability there are bilateral latency measurements between each pair of connected autonomous systems. Latency measurement results are stored in distributed decentralized blockchain network where each AS holds its share. Each AS hosts an SDN instance for interconnection with neighbor SDN controllers over AS border. SDN instances processes latency measurement data from blockchain network, validates data integrity and manages latency aware routing for real-time data flows.","PeriodicalId":198244,"journal":{"name":"2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133872722","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 : 2018-11-01DOI: 10.1109/AIEEE.2018.8592464
T. Sledevič
The article presents the classification of images with pollen bearing bees using convolutional neural network (CNN). The aim is to find out a sufficient configuration of CNN required for future implementation on low-cost FPGA. A new dataset with bee images was collected on the entrances to several beehives. hidden layers with up to 15 15 x 15 down to 3x3 filter sizes. The CNN configured to three hidden layers 7–7, 5–5, 3–3 was selected for future application as a trade off between accuracy 94% and number of required arithmetic operations.
本文介绍了卷积神经网络(CNN)对花粉蜜蜂图像的分类。目的是找出未来在低成本FPGA上实现所需的足够的CNN配置。在几个蜂箱的入口处收集了一个新的蜜蜂图像数据集。隐藏层与多达15 15 x 15到3x3过滤器尺寸。配置为3个隐藏层(7 - 7,5 - 5,3 - 3)的CNN被选择用于未来的应用，以在94%的准确率和所需的算术运算次数之间进行权衡。
{"title":"The Application of Convolutional Neural Network for Pollen Bearing Bee Classification","authors":"T. Sledevič","doi":"10.1109/AIEEE.2018.8592464","DOIUrl":"https://doi.org/10.1109/AIEEE.2018.8592464","url":null,"abstract":"The article presents the classification of images with pollen bearing bees using convolutional neural network (CNN). The aim is to find out a sufficient configuration of CNN required for future implementation on low-cost FPGA. A new dataset with bee images was collected on the entrances to several beehives. hidden layers with up to 15 15 x 15 down to 3x3 filter sizes. The CNN configured to three hidden layers 7–7, 5–5, 3–3 was selected for future application as a trade off between accuracy 94% and number of required arithmetic operations.","PeriodicalId":198244,"journal":{"name":"2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133127084","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 : 2018-11-01DOI: 10.1109/AIEEE.2018.8592423
Linas Petkevičius, R. Baronas
The paper presents a non-linear mathematical model of the batch stirred tank reactor with spherical catalyst particles as microreactors. The different initial conditions are investigated. The influence of the diffusion modulus, the Biot number and the inverse adsorption capacity on the process duration has been numerically investigated. The simulation results showed that process durations might be different by magnitude of order just by different initial microreactor injections. The digital simulation was carried out using the finite difference technique.
{"title":"Numerical Analysis of the Dynamics of Reactant Conversion in Batch Stirred Tank Reactor","authors":"Linas Petkevičius, R. Baronas","doi":"10.1109/AIEEE.2018.8592423","DOIUrl":"https://doi.org/10.1109/AIEEE.2018.8592423","url":null,"abstract":"The paper presents a non-linear mathematical model of the batch stirred tank reactor with spherical catalyst particles as microreactors. The different initial conditions are investigated. The influence of the diffusion modulus, the Biot number and the inverse adsorption capacity on the process duration has been numerically investigated. The simulation results showed that process durations might be different by magnitude of order just by different initial microreactor injections. The digital simulation was carried out using the finite difference technique.","PeriodicalId":198244,"journal":{"name":"2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117236452","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 : 2018-11-01DOI: 10.1109/AIEEE.2018.8592452
A. Jasevičs, L. Zemite, L. Petrichenko
The European Union target for the year 2030 is to achieve a renewable energy source share of at least 32% in the end consumption. The share of electricity generated from renewable energy sources (RES) is increasing every year. In order to stimulate an increase in the intensity of RES and particularly photovoltaic (PV) technology utilisation in countries with a low average solar irradiance, it is necessary to apply various support schemes to minimise the payback period of the technologies. The study includes an evaluation of the current and potential situation regarding the utilisation of PV technology in order for countries with a low average annual solar irradiance to reach the PV technology utilisation level observed in the EU leading countries. The results show that PV technologies make a great contribution to achieving RES goals, however, an increase in their utilisation is closely linked to the amount of investment and to government support to the introduction of the technologies.
{"title":"A Comparative Assessment of the Deployment of PV Technologies in the Baltics and in the European Union","authors":"A. Jasevičs, L. Zemite, L. Petrichenko","doi":"10.1109/AIEEE.2018.8592452","DOIUrl":"https://doi.org/10.1109/AIEEE.2018.8592452","url":null,"abstract":"The European Union target for the year 2030 is to achieve a renewable energy source share of at least 32% in the end consumption. The share of electricity generated from renewable energy sources (RES) is increasing every year. In order to stimulate an increase in the intensity of RES and particularly photovoltaic (PV) technology utilisation in countries with a low average solar irradiance, it is necessary to apply various support schemes to minimise the payback period of the technologies. The study includes an evaluation of the current and potential situation regarding the utilisation of PV technology in order for countries with a low average annual solar irradiance to reach the PV technology utilisation level observed in the EU leading countries. The results show that PV technologies make a great contribution to achieving RES goals, however, an increase in their utilisation is closely linked to the amount of investment and to government support to the introduction of the technologies.","PeriodicalId":198244,"journal":{"name":"2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130022342","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 : 2018-11-01DOI: 10.1109/AIEEE.2018.8592368
P. Suskis, Denis Nikonorov
Most of modern converters are controlled in discrete time domain with digital converter. Therefore, accurate enough plant model is required for a design of the digital control system. A paper contains common approach on digital voltage mode PI controller design for calculated plant model. In comparison to boost, the buck converter has no RHP and performs with high linearity in CCM, the real converter may have differences. A low gain perturbance at different frequencies may be introduced to a PWM controller to measure gain and phase reaction on a small-signal change to get a Bode-plot of a plant. Estimated and calculated plants are compared and analyzed to achieve better performance in real device.
{"title":"Buck Converter Digital Control System Design and Performance Evaluation","authors":"P. Suskis, Denis Nikonorov","doi":"10.1109/AIEEE.2018.8592368","DOIUrl":"https://doi.org/10.1109/AIEEE.2018.8592368","url":null,"abstract":"Most of modern converters are controlled in discrete time domain with digital converter. Therefore, accurate enough plant model is required for a design of the digital control system. A paper contains common approach on digital voltage mode PI controller design for calculated plant model. In comparison to boost, the buck converter has no RHP and performs with high linearity in CCM, the real converter may have differences. A low gain perturbance at different frequencies may be introduced to a PWM controller to measure gain and phase reaction on a small-signal change to get a Bode-plot of a plant. Estimated and calculated plants are compared and analyzed to achieve better performance in real device.","PeriodicalId":198244,"journal":{"name":"2018 IEEE 6th Workshop on Advances in Information, Electronic and Electrical Engineering (AIEEE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116120033","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：