2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)最新文献
Pub Date : 2021-11-24DOI: 10.1109/SNPD51163.2021.9704884
Jiaqing Jian, Chuin-Mu Wang
Many people believe that when drowning occurs, there will be calls for help. In fact, people who are drowning do not get too many splashes or cry for help. They only try to get themselves out of the water by treading on the water. The drowning condition may cause serious brain damage, so it is extremely important to shorten the time it takes to detect the occurrence of drowning and rescue.This paper proposes using computer image processing technology to introduce artificial intelligence motion technology, mounting the camera on the bottom of the swimming pool, and use OpenPose to mark the image joint point features, and input the captured joint point features into the recursive neural network to determine whether the swimmer is drowning. The final training result is about 89.4% accurate, so it can be used to assist on-site lifeguards to detect swimmers who may be drowning, and to reduce incidents that cannot be detected immediately
{"title":"Deep Learning Used to Recognition Swimmers Drowning","authors":"Jiaqing Jian, Chuin-Mu Wang","doi":"10.1109/SNPD51163.2021.9704884","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9704884","url":null,"abstract":"Many people believe that when drowning occurs, there will be calls for help. In fact, people who are drowning do not get too many splashes or cry for help. They only try to get themselves out of the water by treading on the water. The drowning condition may cause serious brain damage, so it is extremely important to shorten the time it takes to detect the occurrence of drowning and rescue.This paper proposes using computer image processing technology to introduce artificial intelligence motion technology, mounting the camera on the bottom of the swimming pool, and use OpenPose to mark the image joint point features, and input the captured joint point features into the recursive neural network to determine whether the swimmer is drowning. The final training result is about 89.4% accurate, so it can be used to assist on-site lifeguards to detect swimmers who may be drowning, and to reduce incidents that cannot be detected immediately","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121072784","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 : 2021-11-24DOI: 10.1109/SNPD51163.2021.9704965
K. Tan, F. Lin
A novel method is proposed to compensate the three-phase unbalanced currents of power grid under three-phase unbalanced load for a two-stage photovoltaic (PV) power system without the augmentation of active power filter (APF). The PV power system is composed of an interleaved DC/DC converter and a three-level neutral-point clamped (NPC) inverter. Moreover, in the proposed method, dq0-axis compensation currents are obtained through low pass filters (LPFs) to compensate the three-phase unbalanced currents of power grid. Furthermore, to improve the control performance of the DC bus voltage of the PV power system under unbalanced load variation condition, an online trained compensatory neural fuzzy network with an asymmetric membership function (CFNN-AMF) is proposed to replace the traditional proportional-integral (PI) controller for the DC bus voltage control. In the proposed CFNN-AMF, the compensatory parameter to integrate pessimistic and optimistic operations of fuzzy systems is embedded in the CFNN. In addition, the dimensions of the Gaussian membership functions are directly extended to AMFs. Additionally, the proposed controllers of the PV power system are implemented by two control platforms using floating-point digital signal processor (DSP). Finally, excellent compensation performance for the three-phase currents of power grid under three-phase unbalanced load can be achieved from the experimental results.
{"title":"PV System Using Intelligent Controller for Unbalanced Current Compensation","authors":"K. Tan, F. Lin","doi":"10.1109/SNPD51163.2021.9704965","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9704965","url":null,"abstract":"A novel method is proposed to compensate the three-phase unbalanced currents of power grid under three-phase unbalanced load for a two-stage photovoltaic (PV) power system without the augmentation of active power filter (APF). The PV power system is composed of an interleaved DC/DC converter and a three-level neutral-point clamped (NPC) inverter. Moreover, in the proposed method, dq0-axis compensation currents are obtained through low pass filters (LPFs) to compensate the three-phase unbalanced currents of power grid. Furthermore, to improve the control performance of the DC bus voltage of the PV power system under unbalanced load variation condition, an online trained compensatory neural fuzzy network with an asymmetric membership function (CFNN-AMF) is proposed to replace the traditional proportional-integral (PI) controller for the DC bus voltage control. In the proposed CFNN-AMF, the compensatory parameter to integrate pessimistic and optimistic operations of fuzzy systems is embedded in the CFNN. In addition, the dimensions of the Gaussian membership functions are directly extended to AMFs. Additionally, the proposed controllers of the PV power system are implemented by two control platforms using floating-point digital signal processor (DSP). Finally, excellent compensation performance for the three-phase currents of power grid under three-phase unbalanced load can be achieved from the experimental results.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125887817","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 : 2021-11-24DOI: 10.1109/SNPD51163.2021.9704956
Shaohui Liu, Shih-Yen Huang
Location failure is dangerous for self-driving vehicles. Adaptive Monte Carlo Localization (AMCL)[1] provides wrong coordinates to the self-driving controller in some specific conditions. This paper proposed a scheme to solve this problem. This scheme provides a reference location to AMCL, which could exactly give coordinates to the self-driving controller. The experiment results showed that this reference location could improve the performance of AMCL to provide precise coordinates to the self-driving controller. In addition, to provide reference location to AMCL, this proposed scheme applied Convolutional Neural Network (CNN)[2] to identify the specific scenery front the vehicle. Accordingly, detect particular views will be another challenge for self-driving vehicles.
{"title":"Apply Image Identification to Improve the Localization of the Self-Driving Vehicles","authors":"Shaohui Liu, Shih-Yen Huang","doi":"10.1109/SNPD51163.2021.9704956","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9704956","url":null,"abstract":"Location failure is dangerous for self-driving vehicles. Adaptive Monte Carlo Localization (AMCL)[1] provides wrong coordinates to the self-driving controller in some specific conditions. This paper proposed a scheme to solve this problem. This scheme provides a reference location to AMCL, which could exactly give coordinates to the self-driving controller. The experiment results showed that this reference location could improve the performance of AMCL to provide precise coordinates to the self-driving controller. In addition, to provide reference location to AMCL, this proposed scheme applied Convolutional Neural Network (CNN)[2] to identify the specific scenery front the vehicle. Accordingly, detect particular views will be another challenge for self-driving vehicles.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128419685","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 : 2021-11-24DOI: 10.1109/SNPD51163.2021.9704970
Chi-Chun Chen, Shang-Lin Tien, Yanhui Lin, Chung-Chen Teng, Meng-Hua Yen
Eco-driving is an effective and immediate environmental protection and energy saving method. This research assists occupational driving license training to achieve eco-driving at two parts: 1. Combine g-sensor with on board diagnostics (OBD-II) and add parameters to improve the data analysis. 2. Through two kinds of neural network models, predict fuel consumption to analyze driving style, and provide reports to display evaluation and behavior suggestions. The experimental configuration designed in this research includes user interface, OBD-II system, neural network model, and is applied to public institutions to provide assistance. The results of this study show that the accuracy of predicting fuel consumption exceeds 97%, which verifies the practicability of the system. The system will also help extend other related applications, such as achieving a driving behavior model that compares energy saving and safety.
{"title":"Truck Driving Assistance System","authors":"Chi-Chun Chen, Shang-Lin Tien, Yanhui Lin, Chung-Chen Teng, Meng-Hua Yen","doi":"10.1109/SNPD51163.2021.9704970","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9704970","url":null,"abstract":"Eco-driving is an effective and immediate environmental protection and energy saving method. This research assists occupational driving license training to achieve eco-driving at two parts: 1. Combine g-sensor with on board diagnostics (OBD-II) and add parameters to improve the data analysis. 2. Through two kinds of neural network models, predict fuel consumption to analyze driving style, and provide reports to display evaluation and behavior suggestions. The experimental configuration designed in this research includes user interface, OBD-II system, neural network model, and is applied to public institutions to provide assistance. The results of this study show that the accuracy of predicting fuel consumption exceeds 97%, which verifies the practicability of the system. The system will also help extend other related applications, such as achieving a driving behavior model that compares energy saving and safety.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133617035","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 : 2021-11-24DOI: 10.1109/SNPD51163.2021.9704963
Ya-Fen Chang, Huan-Wen Chen, TingMao Chang, W. Tai
Recently, Gupta and Chaudhari proposed an anonymous two factor authentication protocol for roaming service in global mobile networks. They claimed that their scheme could not only ensure strong user anonymity, mutual authentication and perfect forward secrecy but also resist desynchronization attack, password guessing attack, replay attack, and insider attack. After analyzing their scheme, we find that it suffers from some flaws. First, the foreign agent cannot determine who the home agent is and whether the received request is for itself or not. Second, some operation cannot be executed by the home agent to record the number of authentication failure. Third, the foreign agent cannot determine whether the message received sent by the home agent is for itself or not. Fourth, a malicious user can mount parallel attack to obtain the unauthorized service. In this paper, we will show how these flaws threaten Gupta and Chaudhari’s protocol.
{"title":"Security Analyses of an Anonymous Two Factor Authentication Protocol for Roaming Service in Global Mobile Networks","authors":"Ya-Fen Chang, Huan-Wen Chen, TingMao Chang, W. Tai","doi":"10.1109/SNPD51163.2021.9704963","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9704963","url":null,"abstract":"Recently, Gupta and Chaudhari proposed an anonymous two factor authentication protocol for roaming service in global mobile networks. They claimed that their scheme could not only ensure strong user anonymity, mutual authentication and perfect forward secrecy but also resist desynchronization attack, password guessing attack, replay attack, and insider attack. After analyzing their scheme, we find that it suffers from some flaws. First, the foreign agent cannot determine who the home agent is and whether the received request is for itself or not. Second, some operation cannot be executed by the home agent to record the number of authentication failure. Third, the foreign agent cannot determine whether the message received sent by the home agent is for itself or not. Fourth, a malicious user can mount parallel attack to obtain the unauthorized service. In this paper, we will show how these flaws threaten Gupta and Chaudhari’s protocol.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129319270","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 : 2021-11-24DOI: 10.1109/SNPD51163.2021.9704917
Tomoki Koguchi, Akinori Ihara
Open Source Software (OSS) developers are free to contribute and free to leave a project, if the project is (not) suitable for them. On the one hand, OSS projects need to manage the human resource to continuously maintain OSS in the future. Some existing studies proposed an approach that estimates how long developers contribute to OSS projects. Using developers’ contributions during the first few months in the target project, the proposed model identified long-term contributors or core developers. However, the approach frequently miss to find capable developers because many developers leave the project soon after participating. To avoid the loss of capable developers, this study build a prediction model to identify future active developers based on their past contributions to any OSS projects. Using dataset from four large-scale OSS projects as a case study, we evaluated our proposed model to identify future active developers based on their past contributions to any OSS projects before participating in a future target project. Our proposed approach contributes to manage human resource in OSS development process.
{"title":"Early Identification of Active Developers Based on their Past Contributions in OSS Projects","authors":"Tomoki Koguchi, Akinori Ihara","doi":"10.1109/SNPD51163.2021.9704917","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9704917","url":null,"abstract":"Open Source Software (OSS) developers are free to contribute and free to leave a project, if the project is (not) suitable for them. On the one hand, OSS projects need to manage the human resource to continuously maintain OSS in the future. Some existing studies proposed an approach that estimates how long developers contribute to OSS projects. Using developers’ contributions during the first few months in the target project, the proposed model identified long-term contributors or core developers. However, the approach frequently miss to find capable developers because many developers leave the project soon after participating. To avoid the loss of capable developers, this study build a prediction model to identify future active developers based on their past contributions to any OSS projects. Using dataset from four large-scale OSS projects as a case study, we evaluated our proposed model to identify future active developers based on their past contributions to any OSS projects before participating in a future target project. Our proposed approach contributes to manage human resource in OSS development process.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127965935","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}
In the past, the height of objects was mostly measured by contact methods, and it may damage the objects. Fringe projection, as a new emerging measurement, might effectively improve the problem with the advantages of non-contact, non-destructive, real-time, full-field, and simple installation. This paper mainly discusses the fringe projection measurement technology to measure the height of the objects accurately. With the application of moiré projection, this paper explains the operation mode and theory of the fringe projection system, and then introduces the phase extraction and phase expansion of the sine wave fringe in sequence. With the longitudinal depth correction, a polynomial fitting method is used to establish a "phase-depth relationship", which is applied to the fringe phase to improve the depth measurement accuracy.
{"title":"Determination of the Height of 3D Objects by Moire Measurement","authors":"Shang-Ya Wu, Hsia-Ping Lan, Chofan Hsieh, Kao-Chi Lin, Pin-Yu Yeh, Cheng-Yu Peng","doi":"10.1109/SNPD51163.2021.9704929","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9704929","url":null,"abstract":"In the past, the height of objects was mostly measured by contact methods, and it may damage the objects. Fringe projection, as a new emerging measurement, might effectively improve the problem with the advantages of non-contact, non-destructive, real-time, full-field, and simple installation. This paper mainly discusses the fringe projection measurement technology to measure the height of the objects accurately. With the application of moiré projection, this paper explains the operation mode and theory of the fringe projection system, and then introduces the phase extraction and phase expansion of the sine wave fringe in sequence. With the longitudinal depth correction, a polynomial fitting method is used to establish a \"phase-depth relationship\", which is applied to the fringe phase to improve the depth measurement accuracy.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134522139","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 : 2021-11-24DOI: 10.1109/SNPD51163.2021.9704928
Tien-Yang Hsu, Yu Lu, Tung-Hung Hsieh, Chou-Chen Wang
High efficiency video coding (HEVC) is a very popular video coding standard. The HEVC can achieve high coding efficiency with a lower bitrate for intra frame coding. However, it still needs many bits to finish best rate-distortion (R-D) curve. Since there are only 35 directions prediction modes provided in intra prediction module (IPM), HEVC occurs a large distortion when the image contents are out of these prediction directions. In order to obtain a better R-D curve, Zhang et al. [3] recently proposed a simple convolutional neural network (S-CNN) to improve the encoding performance of HEVC. However, S-CNN has to consume more time to encode intra frame coding since it needs to perform more CNN enhancement mode. In order to further speed up S-CNN based intra frame coding, we propose an early termination algorithm to skip CNN. Because the natural images are generally homogenous, we find the mean square errors (MSE) of reconstructed CTU exist high spatial correlation at HEVC encoder. Therefore, a dynamic threshold of MSE is set according to three neighboring encoded CTU blocks to evaluate whether the current reconstructed CTU is useful for the CNN enhancement mode. Simulation results show that the proposed method can achieve faster HEVC encoding process than S-CNN by reducing time increase ratio (TIR) about 12% on an average.
{"title":"An Efficient HEVC Intra Frame Coding Based on Deep Convolutional Neural Network","authors":"Tien-Yang Hsu, Yu Lu, Tung-Hung Hsieh, Chou-Chen Wang","doi":"10.1109/SNPD51163.2021.9704928","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9704928","url":null,"abstract":"High efficiency video coding (HEVC) is a very popular video coding standard. The HEVC can achieve high coding efficiency with a lower bitrate for intra frame coding. However, it still needs many bits to finish best rate-distortion (R-D) curve. Since there are only 35 directions prediction modes provided in intra prediction module (IPM), HEVC occurs a large distortion when the image contents are out of these prediction directions. In order to obtain a better R-D curve, Zhang et al. [3] recently proposed a simple convolutional neural network (S-CNN) to improve the encoding performance of HEVC. However, S-CNN has to consume more time to encode intra frame coding since it needs to perform more CNN enhancement mode. In order to further speed up S-CNN based intra frame coding, we propose an early termination algorithm to skip CNN. Because the natural images are generally homogenous, we find the mean square errors (MSE) of reconstructed CTU exist high spatial correlation at HEVC encoder. Therefore, a dynamic threshold of MSE is set according to three neighboring encoded CTU blocks to evaluate whether the current reconstructed CTU is useful for the CNN enhancement mode. Simulation results show that the proposed method can achieve faster HEVC encoding process than S-CNN by reducing time increase ratio (TIR) about 12% on an average.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130129697","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 : 2021-11-24DOI: 10.1109/SNPD51163.2021.9704935
Wei-Chen Li, Ting-Hsuan Hsu, Ke-Nung Huang, Chou-Chen Wang
In recent years, automatic license plate recognition (ALPR) system is applied in some traffic-related applications based on deep learning. However, the new ALPR is very difficult to obtain high detection and recognition rates for oblique car license plate (LP). Recently, Silva et al. [5] proposed a warped planar object detection (WPOD) based on deep convolutional neural network (CNN) to overcome the oblique views of LP. Although the WPOD network can achieve the location and rectification of LPs, the loss function of WPOD renders the confidence parameter due to high computational complexity. This also leads to WPOD network cannot locate the optimal LP bounding box. In order to further improve the accuracy of ALPR system, we develop a simple intersection over union (IOU) algorithm to speed up the calculating process of confidence. In this paper, the four-vertex coordinates of the label bounding box and prediction bounding box of oblique LP are used to generate two rectangular boxes, and then a simple IOU algorithm is used to fast calculate the approximate value of IOU. Simulation results show that the proposed ALPR system can arrive a high accuracy of LP recognition about 95.7% on an average. In addition, the proposed system also can achieve higher recognition rate about 1% when compared to the Silva’s ALPR system.
{"title":"A YOLO-Based Method for Oblique Car License Plate Detection and Recognition","authors":"Wei-Chen Li, Ting-Hsuan Hsu, Ke-Nung Huang, Chou-Chen Wang","doi":"10.1109/SNPD51163.2021.9704935","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9704935","url":null,"abstract":"In recent years, automatic license plate recognition (ALPR) system is applied in some traffic-related applications based on deep learning. However, the new ALPR is very difficult to obtain high detection and recognition rates for oblique car license plate (LP). Recently, Silva et al. [5] proposed a warped planar object detection (WPOD) based on deep convolutional neural network (CNN) to overcome the oblique views of LP. Although the WPOD network can achieve the location and rectification of LPs, the loss function of WPOD renders the confidence parameter due to high computational complexity. This also leads to WPOD network cannot locate the optimal LP bounding box. In order to further improve the accuracy of ALPR system, we develop a simple intersection over union (IOU) algorithm to speed up the calculating process of confidence. In this paper, the four-vertex coordinates of the label bounding box and prediction bounding box of oblique LP are used to generate two rectangular boxes, and then a simple IOU algorithm is used to fast calculate the approximate value of IOU. Simulation results show that the proposed ALPR system can arrive a high accuracy of LP recognition about 95.7% on an average. In addition, the proposed system also can achieve higher recognition rate about 1% when compared to the Silva’s ALPR system.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129957142","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 : 2021-11-24DOI: 10.1109/SNPD51163.2021.9705000
Ayushman Singh, Kaustuv Nag
Here, we propose a structure-preserving deep autoencoder-based dimensionality reduction scheme for data visualization. For this, we introduce two regularizers for regularizing autoencoders. The proposed regularizers help the encoded feature space preserve the local and global structures present in the original feature space. A chosen reduced dimensionality of two or three for the encoded feature space enables us to visualize the extracted latent representations of the data using scatterplots. The proposed method has two variants, depending on which regularizer it uses. The proposed approach, moreover, is unsupervised and has predictability. We use three synthetic datasets and one real-world dataset to illustrate the effectiveness of the proposed method. We also visually compare it with three state-of-the-art data visualization schemes and discuss several future research directions.
{"title":"Structure-Preserving Deep Autoencoder-based Dimensionality Reduction for Data Visualization","authors":"Ayushman Singh, Kaustuv Nag","doi":"10.1109/SNPD51163.2021.9705000","DOIUrl":"https://doi.org/10.1109/SNPD51163.2021.9705000","url":null,"abstract":"Here, we propose a structure-preserving deep autoencoder-based dimensionality reduction scheme for data visualization. For this, we introduce two regularizers for regularizing autoencoders. The proposed regularizers help the encoded feature space preserve the local and global structures present in the original feature space. A chosen reduced dimensionality of two or three for the encoded feature space enables us to visualize the extracted latent representations of the data using scatterplots. The proposed method has two variants, depending on which regularizer it uses. The proposed approach, moreover, is unsupervised and has predictability. We use three synthetic datasets and one real-world dataset to illustrate the effectiveness of the proposed method. We also visually compare it with three state-of-the-art data visualization schemes and discuss several future research directions.","PeriodicalId":235370,"journal":{"name":"2021 IEEE/ACIS 22nd International Conference on Software Engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2021-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133704507","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}