Pub Date : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535588
Y. Paranchuk, Oleksiy Kuznyetsov, V. Tsyapa, Ihor Bilyakovskyy
The structure of an electromechanical system (EMS) with the positioning control loop operating based on fuzzy PI position controller is developed. The model of adaptation of the positioning process to the change of position reference signal is proposed. The solution allows us to obtain optimal plant’s laws of motion in the full positioning range in different directions. The structure of a fuzzy proportional-plus-integral (FPI) position controller based on the Mamdani algorithm is proposed and designed. The structural Simulink model of the positioning EMS with the adaptive fuzzy PI controller is developed, and computer simulations of the positioning control for both directions of the plant’s motion are performed. The analysis of the obtained positioning control proved the correctness and effectiveness of the proposed adaptation model, as far as it allows obtaining optimal (without overshoots and dragging modes) plant’s motions into an arbitrary position from both directions.
{"title":"Positioning Electromechanical System with Adaptive Fuzzy Proportional-Plus-Integral Position Controller","authors":"Y. Paranchuk, Oleksiy Kuznyetsov, V. Tsyapa, Ihor Bilyakovskyy","doi":"10.1109/EUROCON52738.2021.9535588","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535588","url":null,"abstract":"The structure of an electromechanical system (EMS) with the positioning control loop operating based on fuzzy PI position controller is developed. The model of adaptation of the positioning process to the change of position reference signal is proposed. The solution allows us to obtain optimal plant’s laws of motion in the full positioning range in different directions. The structure of a fuzzy proportional-plus-integral (FPI) position controller based on the Mamdani algorithm is proposed and designed. The structural Simulink model of the positioning EMS with the adaptive fuzzy PI controller is developed, and computer simulations of the positioning control for both directions of the plant’s motion are performed. The analysis of the obtained positioning control proved the correctness and effectiveness of the proposed adaptation model, as far as it allows obtaining optimal (without overshoots and dragging modes) plant’s motions into an arbitrary position from both directions.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115392458","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 : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535575
Andrii Prasolov, S. Stirenko, Yuri G. Gordienko
The modern methods and architectures for image super resolution which are based on deep neural networks (DNNs) are considered. Several ways of their improvements were proposed and demonstrated. It was shown that the perception models built on MobileNet and EfficientNet families of DNNs turned out to be faster in training and have a better perception loss rate than previously used VGG family. In the more general context the usage of the smaller DNNs with the higher performance and lower size allow researchers to use and deploy them on devices with the limited computational resources for Edge Computing layer.
{"title":"Improvement of Image Super Resolution by Deep Neural Networks","authors":"Andrii Prasolov, S. Stirenko, Yuri G. Gordienko","doi":"10.1109/EUROCON52738.2021.9535575","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535575","url":null,"abstract":"The modern methods and architectures for image super resolution which are based on deep neural networks (DNNs) are considered. Several ways of their improvements were proposed and demonstrated. It was shown that the perception models built on MobileNet and EfficientNet families of DNNs turned out to be faster in training and have a better perception loss rate than previously used VGG family. In the more general context the usage of the smaller DNNs with the higher performance and lower size allow researchers to use and deploy them on devices with the limited computational resources for Edge Computing layer.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115409545","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 : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535593
Yaroslava Pushkarova, V. Panchenko, Y. Kholin
This paper presents an application of the artificial neural network methodology to prediction of solubilities of 1-1 electrolytes in nonaqueous solvents and solvent mixtures, using experimental data available in the literature. It is demonstrated that that the fundamental expressions proposed previously to describe correlations of solubility with physical-chemical properties of solvents, as well as common regression equations, exhibit large deviations and are not suitable for the description and prediction of solubility for a wide range of individual and mixed solvents. In comparison, the radial basis function artificial neural network algorithm is capable of reproducing the solubilities of such common salts as NaI, CsClO4, NaCl and NaBr in a variety of nonaqueous solvents and solvent mixtures. Having used a training set to obtain the fitting coefficients, we are able to calculate accurately the solubilities of the 1-1 electrolytes in other mixtures of nonaqueous solvents. The reported results make it possible to predict solubilities of 1-1 electrolytes in mixed solvents without the need for additional experimental measurements.
{"title":"Application an Artificial Neural Network for Prediction of Substances Solubility","authors":"Yaroslava Pushkarova, V. Panchenko, Y. Kholin","doi":"10.1109/EUROCON52738.2021.9535593","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535593","url":null,"abstract":"This paper presents an application of the artificial neural network methodology to prediction of solubilities of 1-1 electrolytes in nonaqueous solvents and solvent mixtures, using experimental data available in the literature. It is demonstrated that that the fundamental expressions proposed previously to describe correlations of solubility with physical-chemical properties of solvents, as well as common regression equations, exhibit large deviations and are not suitable for the description and prediction of solubility for a wide range of individual and mixed solvents. In comparison, the radial basis function artificial neural network algorithm is capable of reproducing the solubilities of such common salts as NaI, CsClO4, NaCl and NaBr in a variety of nonaqueous solvents and solvent mixtures. Having used a training set to obtain the fitting coefficients, we are able to calculate accurately the solubilities of the 1-1 electrolytes in other mixtures of nonaqueous solvents. The reported results make it possible to predict solubilities of 1-1 electrolytes in mixed solvents without the need for additional experimental measurements.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116924708","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 : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535613
N. Yeliseyeva, S. Berdnik, V. Katrich
On the base of the solving of the 3-D vector problem of diffraction of the fields of two impedance monopoles located on a perfectly conducting rectangular screen the fast active software for calculating directive gain in the maximum radiation Dmax of two impedance monopoles are developed. There have been used the uniform asymptotics for diffracted fields with account the secondary diffraction on the screen edges and the asymptotic for electric current of a thin impedance dipole of finite length located in free space. The directive gain is investigated vs. the distance between the monopoles, the size L and screen sides ratio W/L. It is shown that at the found optimal distance between the monopoles 0.65λ and optimal screen sizes the directive gain can be increased three times in comparison with the minimum value Dmax. in a given range of screen sizes.
{"title":"Optimization of Directive Gain of Two Impedance Monopoles Located on Metal Rectangular Screen","authors":"N. Yeliseyeva, S. Berdnik, V. Katrich","doi":"10.1109/EUROCON52738.2021.9535613","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535613","url":null,"abstract":"On the base of the solving of the 3-D vector problem of diffraction of the fields of two impedance monopoles located on a perfectly conducting rectangular screen the fast active software for calculating directive gain in the maximum radiation Dmax of two impedance monopoles are developed. There have been used the uniform asymptotics for diffracted fields with account the secondary diffraction on the screen edges and the asymptotic for electric current of a thin impedance dipole of finite length located in free space. The directive gain is investigated vs. the distance between the monopoles, the size L and screen sides ratio W/L. It is shown that at the found optimal distance between the monopoles 0.65λ and optimal screen sizes the directive gain can be increased three times in comparison with the minimum value Dmax. in a given range of screen sizes.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"-1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127136094","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 : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535587
Cheng Wang, Benjamin D. Bowes, P. Beling, J. Goodall
Compared with capital improvement projects, real-time control of stormwater systems may be a more effective and efficient approach to address the increasing risk of flooding in urban areas. One way to automate the design process of control policies is through reinforcement learning (RL). Recently, RL methods have been applied to small stormwater systems and have demonstrated better performance over passive systems and simple rule-based strategies. However, it remains unclear how effective RL methods are for larger and more complex systems. Current RL-based control policies also suffer from poor convergence and stability, which may be due to large updates made by the underlying RL algorithm. In this study, we use the Proximal Policy Optimization (PPO) algorithm and develop control policies for a medium-sized stormwater system that can significantly mitigate flooding during large storm events. Our approach demonstrates good convergence behavior and stability, and achieves robust out-of-sample performance.
{"title":"Reinforcement Learning for Flooding Mitigation in Complex Stormwater Systems during Large Storms","authors":"Cheng Wang, Benjamin D. Bowes, P. Beling, J. Goodall","doi":"10.1109/EUROCON52738.2021.9535587","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535587","url":null,"abstract":"Compared with capital improvement projects, real-time control of stormwater systems may be a more effective and efficient approach to address the increasing risk of flooding in urban areas. One way to automate the design process of control policies is through reinforcement learning (RL). Recently, RL methods have been applied to small stormwater systems and have demonstrated better performance over passive systems and simple rule-based strategies. However, it remains unclear how effective RL methods are for larger and more complex systems. Current RL-based control policies also suffer from poor convergence and stability, which may be due to large updates made by the underlying RL algorithm. In this study, we use the Proximal Policy Optimization (PPO) algorithm and develop control policies for a medium-sized stormwater system that can significantly mitigate flooding during large storm events. Our approach demonstrates good convergence behavior and stability, and achieves robust out-of-sample performance.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123338054","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 : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535586
Yevhenii Trochun, Evgen Pavlov, S. Stirenko, Yuri G. Gordienko
This article describes hybrid convolutional neural network that uses one quantum circuit for image classification. The different configurations of the hybrid neural network with the quantum circuit are considered. Several different quantum circuits with different number of qubits are compared. These hybrid neural network configurations are evaluated on MNIST and MNIST Fashion datasets, which is radically different from MNIST dataset. Performance of hybrid neural network is compared for multiclass classification on MNIST and MNIST Fashion datasets for 4, 6, 8, 10 classes using quantum circuits with 2, 3, 4 qubits. The results of the experiments indicate the feasibility of using hybrid neural networks for multiclass classification.
{"title":"Impact of Hybrid Neural Network Structure on Performance of Multiclass Classification","authors":"Yevhenii Trochun, Evgen Pavlov, S. Stirenko, Yuri G. Gordienko","doi":"10.1109/EUROCON52738.2021.9535586","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535586","url":null,"abstract":"This article describes hybrid convolutional neural network that uses one quantum circuit for image classification. The different configurations of the hybrid neural network with the quantum circuit are considered. Several different quantum circuits with different number of qubits are compared. These hybrid neural network configurations are evaluated on MNIST and MNIST Fashion datasets, which is radically different from MNIST dataset. Performance of hybrid neural network is compared for multiclass classification on MNIST and MNIST Fashion datasets for 4, 6, 8, 10 classes using quantum circuits with 2, 3, 4 qubits. The results of the experiments indicate the feasibility of using hybrid neural networks for multiclass classification.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123752941","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 : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535565
Y. Priyadarshana, L. Ranathunga, C. Amalraj, I. Perera
Named entity recognition (NER) is a prominent task in identifying text spans to specific types. Named entity boundary detection can be mentioned as a rising research area under NER. Although a limited work has been conducted for nested NE boundary detection, flat NE boundary detection can be considered as at a pinnacle stage. Nested NE boundary detection is an important aspect in information extraction, information retrieval, event extraction, sentiment analysis etc. On the other hand, spreading religious unhealthy statements through social media has become a burden for the wellbeing of the society. The prime objective of this research is to implement a novel system for nested NE boundary detection for Sinhala language considering religious unhealthy statements in social media. A constructive literature survey has been conducted for analyzing the already developed NE type and boundary detection approaches and systems. Along with that, identifying the linguistic structures and patterns of Sinhala hate speech detection has been conducted. A corpus of more than 100,000 Sinhala hates speech contents have been extracted, preprocessed, and annotated by an expert panel. Then, a deep neural approach has been applied for capturing the complexity indexes, matrices, and other related elements of the corpus. Next, a novel approach called "boundary bubbles" has been conducted for capturing word representation, head word detection, entity mention nuggets identification and region classification for NE boundary detection. Experiments reveal that our scientific novel approach has achieved the state-of-art performance over the existing baselines.
{"title":"HelaNER: A Novel Approach for Nested Named Entity Boundary Detection","authors":"Y. Priyadarshana, L. Ranathunga, C. Amalraj, I. Perera","doi":"10.1109/EUROCON52738.2021.9535565","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535565","url":null,"abstract":"Named entity recognition (NER) is a prominent task in identifying text spans to specific types. Named entity boundary detection can be mentioned as a rising research area under NER. Although a limited work has been conducted for nested NE boundary detection, flat NE boundary detection can be considered as at a pinnacle stage. Nested NE boundary detection is an important aspect in information extraction, information retrieval, event extraction, sentiment analysis etc. On the other hand, spreading religious unhealthy statements through social media has become a burden for the wellbeing of the society. The prime objective of this research is to implement a novel system for nested NE boundary detection for Sinhala language considering religious unhealthy statements in social media. A constructive literature survey has been conducted for analyzing the already developed NE type and boundary detection approaches and systems. Along with that, identifying the linguistic structures and patterns of Sinhala hate speech detection has been conducted. A corpus of more than 100,000 Sinhala hates speech contents have been extracted, preprocessed, and annotated by an expert panel. Then, a deep neural approach has been applied for capturing the complexity indexes, matrices, and other related elements of the corpus. Next, a novel approach called \"boundary bubbles\" has been conducted for capturing word representation, head word detection, entity mention nuggets identification and region classification for NE boundary detection. Experiments reveal that our scientific novel approach has achieved the state-of-art performance over the existing baselines.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122304408","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 : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535644
S. Rai, M. De
This paper discusses a load contribution factor (LCF) based multinodal load forecasting technique. The dynamic nature of the electrical load in any distribution network is the reason behind the need for simultaneous forecasting of load at all the nodes. In a small distribution system, the load at different nodes is interdependent to each other, and also all the nodes are located at similar physical locations and hence loads cannot be distinguished based on weather parameters. Due to this, multinodal load forecasting becomes a tough job. This problem is solved by using LCF for training the load forecasting model along with the other exogenous factors. LCF is calculated for each node depending upon the past trend of load at that node at any particular instant and the total load of the system. Results of the proposed method produce accurate and consistent multinodal load forecasting performance for the real-time smart-metered data available at the residential academic campus grid.
{"title":"Effect of Load Contribution Factor on Multinodal Load Forecasting","authors":"S. Rai, M. De","doi":"10.1109/EUROCON52738.2021.9535644","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535644","url":null,"abstract":"This paper discusses a load contribution factor (LCF) based multinodal load forecasting technique. The dynamic nature of the electrical load in any distribution network is the reason behind the need for simultaneous forecasting of load at all the nodes. In a small distribution system, the load at different nodes is interdependent to each other, and also all the nodes are located at similar physical locations and hence loads cannot be distinguished based on weather parameters. Due to this, multinodal load forecasting becomes a tough job. This problem is solved by using LCF for training the load forecasting model along with the other exogenous factors. LCF is calculated for each node depending upon the past trend of load at that node at any particular instant and the total load of the system. Results of the proposed method produce accurate and consistent multinodal load forecasting performance for the real-time smart-metered data available at the residential academic campus grid.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122843358","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 : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535550
Mickael Cormier, Stefan Wolf, L. Sommer, Arne Schumann, J. Beyerer
The behavior of individuals in crowds in public places has gained enormously in importance last year, for example through distancing requirements. However, automatically detecting pedestrians in real-world uncooperative scenarios remains a very challenging task. Especially crowded areas in surveillance footage are not only challenging for automatic vision systems, but also for human operators. Furthermore, complex detection models do not scale easily and are not traditionally designed for on-device processing in resource-constrained smart cameras, which become more and more popular due to technical and privacy issues at large events. In this work, we propose a new Fast Pedestrian Detector (FPD) based on RetinaNet which is a fast and efficient architecture for embedded platforms. The proposed FPD provides near real-time and real-time detection of hundreds of pedestrians on embedded platforms, outperforming popular YOLO-based approaches traditionally tuned for speed. Furthermore, by evaluating our approach on several different Jetson platforms in terms of speed and energy profiles, we highlight the challenges related to the deployment of a deep learning based pedestrian detector on embedded platforms for smart surveillance cameras.
{"title":"Fast Pedestrian Detection for Real-World Crowded Scenarios on Embedded GPU","authors":"Mickael Cormier, Stefan Wolf, L. Sommer, Arne Schumann, J. Beyerer","doi":"10.1109/EUROCON52738.2021.9535550","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535550","url":null,"abstract":"The behavior of individuals in crowds in public places has gained enormously in importance last year, for example through distancing requirements. However, automatically detecting pedestrians in real-world uncooperative scenarios remains a very challenging task. Especially crowded areas in surveillance footage are not only challenging for automatic vision systems, but also for human operators. Furthermore, complex detection models do not scale easily and are not traditionally designed for on-device processing in resource-constrained smart cameras, which become more and more popular due to technical and privacy issues at large events. In this work, we propose a new Fast Pedestrian Detector (FPD) based on RetinaNet which is a fast and efficient architecture for embedded platforms. The proposed FPD provides near real-time and real-time detection of hundreds of pedestrians on embedded platforms, outperforming popular YOLO-based approaches traditionally tuned for speed. Furthermore, by evaluating our approach on several different Jetson platforms in terms of speed and energy profiles, we highlight the challenges related to the deployment of a deep learning based pedestrian detector on embedded platforms for smart surveillance cameras.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117272684","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 : 2021-07-06DOI: 10.1109/EUROCON52738.2021.9535624
Alexander Gutev, C. J. Debono
Occlusions present a significant challenge to successfully track objects in video content, even with state of the art tracking algorithms. In this paper, a new tracking system, which utilizes the additional information provided by 3D video content, is presented. The system incorporates a 3D Kalman Filter coupled with occlusion reasoning, based on segmentation of the depth map, to accurately track an object during an occlusion and after it reemerges. Results show an improvement in robustness, over the state of art, especially in videos where the tracked object’s motion is linear.
{"title":"Improved Object Tracking Throughout Occlusions","authors":"Alexander Gutev, C. J. Debono","doi":"10.1109/EUROCON52738.2021.9535624","DOIUrl":"https://doi.org/10.1109/EUROCON52738.2021.9535624","url":null,"abstract":"Occlusions present a significant challenge to successfully track objects in video content, even with state of the art tracking algorithms. In this paper, a new tracking system, which utilizes the additional information provided by 3D video content, is presented. The system incorporates a 3D Kalman Filter coupled with occlusion reasoning, based on segmentation of the depth map, to accurately track an object during an occlusion and after it reemerges. Results show an improvement in robustness, over the state of art, especially in videos where the tracked object’s motion is linear.","PeriodicalId":328338,"journal":{"name":"IEEE EUROCON 2021 - 19th International Conference on Smart Technologies","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128497266","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：