Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989128
T. Tran, Thi-Uyen-Uyen Nguyen
Wind turbines are the technology that makes good use of wind energy and have achieved many remarkable achievements. In particular, rotor blades are considered the most important part of the wind turbine system. They are responsible for converting the kinetic energy obtained from the wind into mechanical energy to rotate a generator. The electrical power output of the turbine is largely determined by this part. The design features of the rotor such as rotor sweep area, tip speed ratio, number of blades, design blade profiles, chord and blade twist angle distribution, thickness, and blade material will significantly affect turbine performance. In this paper, the research object will be a small horizontal axis wind turbine, with an expected design capacity of 300 W. Thereby, the research will go from the first steps of a design process, calculate the design and select an effective design for the rotor blade, estimate and test the durability limit of the design according to the international standards. The main purpose of the research is to study, analyse aerodynamic characteristics, select and calculate design parameters of wind turbine rotor blades, then compare the results of different design cases (blade profile, tip speed ratio, Reynolds number) to select the most optimal design solution in terms of performance and design cost with the support mainly by QBlade software (based on the blade element momentum theory BEM with wake rotation and loss). Based on the IEC61400-2 design standard for small wind turbines, this research will find out the loads available to a wind turbine in a simple load case for rotor blades under wind load with QBlade's QFEM tool. Use QBlade's QLLT (Nonlinear Lifting Line Simulation) to simulate the behaviour and collect load estimation data of a wind turbine operating in a turbulent wind velocity field.
{"title":"Process for Calculating Design Parameters of a Small Horizontal Axis Wind Turbine of 300W","authors":"T. Tran, Thi-Uyen-Uyen Nguyen","doi":"10.1109/GTSD54989.2022.9989128","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989128","url":null,"abstract":"Wind turbines are the technology that makes good use of wind energy and have achieved many remarkable achievements. In particular, rotor blades are considered the most important part of the wind turbine system. They are responsible for converting the kinetic energy obtained from the wind into mechanical energy to rotate a generator. The electrical power output of the turbine is largely determined by this part. The design features of the rotor such as rotor sweep area, tip speed ratio, number of blades, design blade profiles, chord and blade twist angle distribution, thickness, and blade material will significantly affect turbine performance. In this paper, the research object will be a small horizontal axis wind turbine, with an expected design capacity of 300 W. Thereby, the research will go from the first steps of a design process, calculate the design and select an effective design for the rotor blade, estimate and test the durability limit of the design according to the international standards. The main purpose of the research is to study, analyse aerodynamic characteristics, select and calculate design parameters of wind turbine rotor blades, then compare the results of different design cases (blade profile, tip speed ratio, Reynolds number) to select the most optimal design solution in terms of performance and design cost with the support mainly by QBlade software (based on the blade element momentum theory BEM with wake rotation and loss). Based on the IEC61400-2 design standard for small wind turbines, this research will find out the loads available to a wind turbine in a simple load case for rotor blades under wind load with QBlade's QFEM tool. Use QBlade's QLLT (Nonlinear Lifting Line Simulation) to simulate the behaviour and collect load estimation data of a wind turbine operating in a turbulent wind velocity field.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"191 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131303471","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989259
Khanh-Phong Bui, Hoang-Lam Ngoc Le, Quang-Thang Le, Dinh-Hiep Huynh, Vu-Hoang Tran
Recently, a lot of research and applications regarding autonomous vehicles have been invested in and developed. These applications have many complex scenarios to handle such as lane segmentation, object detection, traffic sign recognition, and steering control prediction. Many methods handle these tasks separately. Despite the excellent performance these methods achieve, processing these tasks one after another takes a longer time than tackling them all at once. So, in this paper, to reduce the inference time of the autonomous driving system, we proposed a multi-task framework to conduct three tasks: lane segmentation, object detection, and traffic sign recognition simultaneously. Our framework is composed of one encoder for feature extraction and two decoders to handle specific tasks. We only use one encoder for multiple tasks because these tasks complement each other, we hope that the information can be shared among these tasks through the single encoder to improve the performance of each task and also to reduce the amount of data required for training. The decoders include a detection decoder and a segmentation decoder. The detection decoder is designed to detect objects and recognize traffic signs. On the other hand, the segmentation decoder is designed to focus solely on the task of separating the drivable area. By testing on the challenging Carla dataset, our model shows that it can achieve better results compared to state-of-the-art methods. Besides, experimental results also show that, compared with solving tasks independently, our framework can achieve similar performance but greatly reduce processing time.
{"title":"The Design of a Multi-Task Learning System for Autonomous Vehicles","authors":"Khanh-Phong Bui, Hoang-Lam Ngoc Le, Quang-Thang Le, Dinh-Hiep Huynh, Vu-Hoang Tran","doi":"10.1109/GTSD54989.2022.9989259","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989259","url":null,"abstract":"Recently, a lot of research and applications regarding autonomous vehicles have been invested in and developed. These applications have many complex scenarios to handle such as lane segmentation, object detection, traffic sign recognition, and steering control prediction. Many methods handle these tasks separately. Despite the excellent performance these methods achieve, processing these tasks one after another takes a longer time than tackling them all at once. So, in this paper, to reduce the inference time of the autonomous driving system, we proposed a multi-task framework to conduct three tasks: lane segmentation, object detection, and traffic sign recognition simultaneously. Our framework is composed of one encoder for feature extraction and two decoders to handle specific tasks. We only use one encoder for multiple tasks because these tasks complement each other, we hope that the information can be shared among these tasks through the single encoder to improve the performance of each task and also to reduce the amount of data required for training. The decoders include a detection decoder and a segmentation decoder. The detection decoder is designed to detect objects and recognize traffic signs. On the other hand, the segmentation decoder is designed to focus solely on the task of separating the drivable area. By testing on the challenging Carla dataset, our model shows that it can achieve better results compared to state-of-the-art methods. Besides, experimental results also show that, compared with solving tasks independently, our framework can achieve similar performance but greatly reduce processing time.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126956395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989226
Nguyen Duy Khanh, Phan Van Quan, Ha Pham Trong Phu, Dang Xuan Ba
Unmanned aerial vehicles (UAV, drones) are becoming one of the key machines/tools of the modern world, especially in military applications. Numerous research work is underway to explore the possibility of using these machines in other applications such as delivery, protecting wildlife, agricultural activities, academy, searching, and rescue missions. Since these machines are unmanned vehicles, their functionalities are completely dependent upon the performances of their control systems. This article presents a control approach to obtain better stabilization in the attitude and altitude of a quadcopter under different disturbance conditions. The controller proposed is contingent on a Linear Quadratic Regulator (LQR) structure with a state observer (SO). The combination of LQR controller and the state observer is used to control and filter out the sensors and systems noises. Therefore, the quadcopter can be controlled to result in expected control performances in numerous different environments. Finally, the effectiveness of the proposed method is demonstrated through simulation results.
{"title":"An Adaptive Pose Controller of 6DOF Quadcopters Using a State Observer","authors":"Nguyen Duy Khanh, Phan Van Quan, Ha Pham Trong Phu, Dang Xuan Ba","doi":"10.1109/GTSD54989.2022.9989226","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989226","url":null,"abstract":"Unmanned aerial vehicles (UAV, drones) are becoming one of the key machines/tools of the modern world, especially in military applications. Numerous research work is underway to explore the possibility of using these machines in other applications such as delivery, protecting wildlife, agricultural activities, academy, searching, and rescue missions. Since these machines are unmanned vehicles, their functionalities are completely dependent upon the performances of their control systems. This article presents a control approach to obtain better stabilization in the attitude and altitude of a quadcopter under different disturbance conditions. The controller proposed is contingent on a Linear Quadratic Regulator (LQR) structure with a state observer (SO). The combination of LQR controller and the state observer is used to control and filter out the sensors and systems noises. Therefore, the quadcopter can be controlled to result in expected control performances in numerous different environments. Finally, the effectiveness of the proposed method is demonstrated through simulation results.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123416894","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989222
Hồ Anh Khoa, Vuong Thi Ngoc Han, Nguyen Phan Thanh, Do Duc Tri
Partial Shading Conditions (PSC) are a common stumbling block to a solar photovoltaic (PV) system's continuous power generation. PSC will reduce the output power of a PV array significantly. This paper proposes a solution for using a Three-level Quasi-Switched Boost T-type Inverter (3L-qSBT2I) single-stage inverter configuration for PV systems with the control algorithms Sine Pulse Width Modulation (SPWM) and Stabilize DC-link voltage algorithm to overcome the voltage drop and output power of the PV system when PSC occurs. The proposed approach will work with all PV connection setups and will eliminate the drawbacks of basic MPPT. The simulation is implemented with the help of PSIM software to demonstrate the accuracy of this strategy.
{"title":"Modified SinPWM Algorithm For PV Inverter In Partial Shading Conditions","authors":"Hồ Anh Khoa, Vuong Thi Ngoc Han, Nguyen Phan Thanh, Do Duc Tri","doi":"10.1109/GTSD54989.2022.9989222","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989222","url":null,"abstract":"Partial Shading Conditions (PSC) are a common stumbling block to a solar photovoltaic (PV) system's continuous power generation. PSC will reduce the output power of a PV array significantly. This paper proposes a solution for using a Three-level Quasi-Switched Boost T-type Inverter (3L-qSBT2I) single-stage inverter configuration for PV systems with the control algorithms Sine Pulse Width Modulation (SPWM) and Stabilize DC-link voltage algorithm to overcome the voltage drop and output power of the PV system when PSC occurs. The proposed approach will work with all PV connection setups and will eliminate the drawbacks of basic MPPT. The simulation is implemented with the help of PSIM software to demonstrate the accuracy of this strategy.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126280986","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989309
Vũ Hải Quân, N. Khoa, Dinh Xuan Thanh
The article analyzed, selected technology and successfully manufactured a simulation model for the operation of the fuel cell system. The design model allows to investigate the influence of the parameters on the performance of the fuel cell assembly. The model test results show that the standard voltage when connected to a load of a pair of batteries is in the range of 0.6 – 0.7V. The output current and voltage are proportional to the number of battery pairs. The performance of the fuel cell assembly depends on the selection of technology, catalytic materials on the proton exchange membrane, a number of parameters: Feed pressure of hydrogen, oxygen, humidity of supply gas… Working mode The Electrolyzer's work in principle will reverse the operation of the fuel cell assembly, the larger the current supply to the water electrolyzer, the efficiency will increase, but the corrosion rate at the electrodes will increase.
{"title":"Design a Fuel Battery Operation Model for a Car Application for Training","authors":"Vũ Hải Quân, N. Khoa, Dinh Xuan Thanh","doi":"10.1109/GTSD54989.2022.9989309","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989309","url":null,"abstract":"The article analyzed, selected technology and successfully manufactured a simulation model for the operation of the fuel cell system. The design model allows to investigate the influence of the parameters on the performance of the fuel cell assembly. The model test results show that the standard voltage when connected to a load of a pair of batteries is in the range of 0.6 – 0.7V. The output current and voltage are proportional to the number of battery pairs. The performance of the fuel cell assembly depends on the selection of technology, catalytic materials on the proton exchange membrane, a number of parameters: Feed pressure of hydrogen, oxygen, humidity of supply gas… Working mode The Electrolyzer's work in principle will reverse the operation of the fuel cell assembly, the larger the current supply to the water electrolyzer, the efficiency will increase, but the corrosion rate at the electrodes will increase.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"43 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114042969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989308
Phi–Long H. Nguyen, V. Pham
In reality, it is desired to reliably detect unauthorized entries for prohibited areas such as teleoperated high voltage power substations. To this end, this paper presents a vision based people detection method in power substations. More than 42,000 images were collected from the COCO dataset, Youtube and a real 220kV power substation for training models. Deep learning models EfficientDet-D1 and YOLOv5-m were employed for transfer learning. Experimental results show that the EfficientDet-D1 and YOLOv5-m can recognize people in 220kV power substations with mAP of 63.2% and 88.6%, respectively.
{"title":"Vision based People Detection in Power Substations","authors":"Phi–Long H. Nguyen, V. Pham","doi":"10.1109/GTSD54989.2022.9989308","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989308","url":null,"abstract":"In reality, it is desired to reliably detect unauthorized entries for prohibited areas such as teleoperated high voltage power substations. To this end, this paper presents a vision based people detection method in power substations. More than 42,000 images were collected from the COCO dataset, Youtube and a real 220kV power substation for training models. Deep learning models EfficientDet-D1 and YOLOv5-m were employed for transfer learning. Experimental results show that the EfficientDet-D1 and YOLOv5-m can recognize people in 220kV power substations with mAP of 63.2% and 88.6%, respectively.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125123675","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989035
Thao Nguyen Da, Li Yimin, Chi Peng, M. Cho, Khanh Nguyen Le Kim, Phuong Nguyen Thanh
Solar power is a clean energy source that has developed quickly with considerable attention. Solar energy is required more accurate predictions, which could be integrated into the power grid. Therefore, this project attempts to improve short-term solar power prediction's accuracy, utilizing the long short-term memory (LSTM) in a deep learning machine. The collected data is acquired from the solar system installed in Kaohsiung city, Taiwan. The historical sequential weather parameter and the collected data from the battery module are utilized as input features for the predicting model. To acquire the optimum performance, hyperparameter optimization is employed to construct the best sequential historical data of the LSTM model. The experiment results are compared with a recurrent neural network (RNN), indicating that the LSTM could predict short-term solar power better.
{"title":"Short-term Solar Power Prediction using Long Short-Term Memory in Solar Plant with Deep Learning Machine","authors":"Thao Nguyen Da, Li Yimin, Chi Peng, M. Cho, Khanh Nguyen Le Kim, Phuong Nguyen Thanh","doi":"10.1109/GTSD54989.2022.9989035","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989035","url":null,"abstract":"Solar power is a clean energy source that has developed quickly with considerable attention. Solar energy is required more accurate predictions, which could be integrated into the power grid. Therefore, this project attempts to improve short-term solar power prediction's accuracy, utilizing the long short-term memory (LSTM) in a deep learning machine. The collected data is acquired from the solar system installed in Kaohsiung city, Taiwan. The historical sequential weather parameter and the collected data from the battery module are utilized as input features for the predicting model. To acquire the optimum performance, hyperparameter optimization is employed to construct the best sequential historical data of the LSTM model. The experiment results are compared with a recurrent neural network (RNN), indicating that the LSTM could predict short-term solar power better.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"202 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131546365","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989205
V. Dao, Phuong Thi Tuyet Nguyen, Thi-Bich-Nga Truong, Hoang-Minh Le, Thi-Tuyet-Dan Bui, Thi-Ngoc-Hieu Phu, Thi-Ngoc-Thu Vu
Unlike terrestrial wireless networks, which employ radio waves to communicate, underwater acoustic networks (UANs) employ acoustic waves to communicate. The long propagation delay and limited bandwidth of UANs pose great challenges for MAC protocol design. But the long propagation delay in underwater acoustic channel permits multiple packets to be “pipelined” concurrently in the underwater channel. To utilize this characteristic of underwater acoustic channel, we propose a slotted FAMA-based MAC protocol (slotted C-FAMA) for underwater acoustic networks in this paper. In this protocol, we use control packets (RTS; CTS, ACK and NACK) for handshaking and time-slotted based to allocate the time for each node to communicate. Every control packet is transmitted from the source node to the destination node at the beginning of each time-slot to avoid collision. The simulations show that the slotted C-FAMA is able to achieve high and stable throughput performance, and increase channel utilization while maintaining low collision rate.
{"title":"A Scheduled and Slotted MAC Protocol for Underwater Acoustic Networks","authors":"V. Dao, Phuong Thi Tuyet Nguyen, Thi-Bich-Nga Truong, Hoang-Minh Le, Thi-Tuyet-Dan Bui, Thi-Ngoc-Hieu Phu, Thi-Ngoc-Thu Vu","doi":"10.1109/GTSD54989.2022.9989205","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989205","url":null,"abstract":"Unlike terrestrial wireless networks, which employ radio waves to communicate, underwater acoustic networks (UANs) employ acoustic waves to communicate. The long propagation delay and limited bandwidth of UANs pose great challenges for MAC protocol design. But the long propagation delay in underwater acoustic channel permits multiple packets to be “pipelined” concurrently in the underwater channel. To utilize this characteristic of underwater acoustic channel, we propose a slotted FAMA-based MAC protocol (slotted C-FAMA) for underwater acoustic networks in this paper. In this protocol, we use control packets (RTS; CTS, ACK and NACK) for handshaking and time-slotted based to allocate the time for each node to communicate. Every control packet is transmitted from the source node to the destination node at the beginning of each time-slot to avoid collision. The simulations show that the slotted C-FAMA is able to achieve high and stable throughput performance, and increase channel utilization while maintaining low collision rate.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124405558","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989248
Nhat Truong Pham, Van Dung Pham, Qui Nguyen-Van, Bao Hung Nguyen, Duc Ngoc Minh Dang, Sy Dzung Nguyen
Scene text detection and recognition are vital challenging tasks in computer vision, which are to detect and recognize sequences of texts in natural scenes. Recently, researchers have investigated a lot of state-of-the-art methods to improve the accuracy and efficiency of text detection and recognition. However, there has been little research on text detection and recognition in natural scenes in Vietnam. In this paper, a deep learning-based empirical investigation of Vietnamese scene text detection and recognition is presented. Firstly, four detection models including differentiable binarization network (DBN), pyramid mask text detector (PMTD), pixel aggregation network (PAN), and Fourier contour embedding network (FCEN), are employed to detect text regions from the images. Then, four text recognition models including convolutional recurrent neural network (CRNN), self-attention text recognition network (SATRN), no-recurrence sequence-to-sequence text recognizer (NRTR), and RobustScanner (RS) are also investigated to recognize the texts. Moreover, data augmentation methods are also applied to enrich data for improving the accuracy and enhancing the performance of scene text detection and recognition. To validate the effectiveness of scene text detection and recognition models, the VinText dataset is employed for evaluation. Empirical results show that PMTD and SATRN achieve the highest scores among the others for text detection and recognition, respectively. For knowledge-sharing, our implementation is publicly available at https://github.com/ThorPham/VN_scene_text_detection_recognition.
{"title":"Vietnamese Scene Text Detection and Recognition using Deep Learning: An Empirical Study","authors":"Nhat Truong Pham, Van Dung Pham, Qui Nguyen-Van, Bao Hung Nguyen, Duc Ngoc Minh Dang, Sy Dzung Nguyen","doi":"10.1109/GTSD54989.2022.9989248","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989248","url":null,"abstract":"Scene text detection and recognition are vital challenging tasks in computer vision, which are to detect and recognize sequences of texts in natural scenes. Recently, researchers have investigated a lot of state-of-the-art methods to improve the accuracy and efficiency of text detection and recognition. However, there has been little research on text detection and recognition in natural scenes in Vietnam. In this paper, a deep learning-based empirical investigation of Vietnamese scene text detection and recognition is presented. Firstly, four detection models including differentiable binarization network (DBN), pyramid mask text detector (PMTD), pixel aggregation network (PAN), and Fourier contour embedding network (FCEN), are employed to detect text regions from the images. Then, four text recognition models including convolutional recurrent neural network (CRNN), self-attention text recognition network (SATRN), no-recurrence sequence-to-sequence text recognizer (NRTR), and RobustScanner (RS) are also investigated to recognize the texts. Moreover, data augmentation methods are also applied to enrich data for improving the accuracy and enhancing the performance of scene text detection and recognition. To validate the effectiveness of scene text detection and recognition models, the VinText dataset is employed for evaluation. Empirical results show that PMTD and SATRN achieve the highest scores among the others for text detection and recognition, respectively. For knowledge-sharing, our implementation is publicly available at https://github.com/ThorPham/VN_scene_text_detection_recognition.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"2023 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131549694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-29DOI: 10.1109/GTSD54989.2022.9989094
D. C. Huynh, Thanh H. Truong, M. Dunnigan, Corina Barbalata
The great developments in AC-DC-AC conversion technology and the connection of renewable energy sources through long distances have resulted in the extensive growth of high voltage direct current (HVDC) systems. The operation of HVDC systems remains a challenge. In particular, the problem of fault location on HVDC transmission lines plays an important role. This paper proposes a fault location technique on an HVDC transmission line which is based on transforming the fault location problem into an optimization problem. Then, a water wave optimization (WWO) algorithm is proposed to solve the above optimization problem for identifying fault locations on the HVDC transmission line. The WWO algorithm-based fault location results are compared with others to validate the proposal.
{"title":"Optimization-based Fault Location on High Voltage Direct Current Transmission Lines","authors":"D. C. Huynh, Thanh H. Truong, M. Dunnigan, Corina Barbalata","doi":"10.1109/GTSD54989.2022.9989094","DOIUrl":"https://doi.org/10.1109/GTSD54989.2022.9989094","url":null,"abstract":"The great developments in AC-DC-AC conversion technology and the connection of renewable energy sources through long distances have resulted in the extensive growth of high voltage direct current (HVDC) systems. The operation of HVDC systems remains a challenge. In particular, the problem of fault location on HVDC transmission lines plays an important role. This paper proposes a fault location technique on an HVDC transmission line which is based on transforming the fault location problem into an optimization problem. Then, a water wave optimization (WWO) algorithm is proposed to solve the above optimization problem for identifying fault locations on the HVDC transmission line. The WWO algorithm-based fault location results are compared with others to validate the proposal.","PeriodicalId":125445,"journal":{"name":"2022 6th International Conference on Green Technology and Sustainable Development (GTSD)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115244267","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：