Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577728
Yufei Han, P. Xu, Qishuang Ma
Switched reluctance motor (SRM) has attracted attention of the motor industry due to its simple structure, high reliability and low cost. However, the inherent shortcoming of SRM, principally the torque ripple, limited its development. In this paper, the direct torque control (DTC) method for four-phase SRM will be researched. The relationship between space flux vector and voltage vector will be analyzed, and the switch on-off state principle will be established. The DTC control strategy is simulated by using Simulink, compared with the widely used traditional strategy -current chopper control (CCC). The simulation shows that DTC for four-phase (8/6) SRM can reduce the torque ripple with 73% compared to the traditional strategy.
{"title":"Torque ripple reduction of four-phase SRM based on DTC method","authors":"Yufei Han, P. Xu, Qishuang Ma","doi":"10.1109/IAEAC.2018.8577728","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577728","url":null,"abstract":"Switched reluctance motor (SRM) has attracted attention of the motor industry due to its simple structure, high reliability and low cost. However, the inherent shortcoming of SRM, principally the torque ripple, limited its development. In this paper, the direct torque control (DTC) method for four-phase SRM will be researched. The relationship between space flux vector and voltage vector will be analyzed, and the switch on-off state principle will be established. The DTC control strategy is simulated by using Simulink, compared with the widely used traditional strategy -current chopper control (CCC). The simulation shows that DTC for four-phase (8/6) SRM can reduce the torque ripple with 73% compared to the traditional strategy.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"42 1","pages":"994-997"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73297032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577878
Siyun Chen, Ye Wang, Jia Yu, Nian-sheng Wang, Yan Yan
Coordinated Multi Point(CoMP) can be employed to reduce interference and improve network throughput in Ultra Dense Network(UDN). However, backhaul capacity is a bottleneck in achieving the goal. Fortunately, proactive caching is a promising technology to alleviate backhaul load. In a cache-enabled system with CoMP, user association is an important problem which can give prominence to its advantage. In this paper, to pursue large network throughput gain at less cost of backhaul load, we formulate user association problem as an Integer Programming(IP) problem. Given the strategy of content placement, we propose an N-Best Binary Particle Swarm Optimization(NB-BPSO) algorithm to solve the problem. Simulation results show that our algorithm improves network throughput and reduces backhaul traffic significantly compared with traditional user association schemes.
{"title":"User Association in Cache-enabled Ultra Dense Network with JT CoMP","authors":"Siyun Chen, Ye Wang, Jia Yu, Nian-sheng Wang, Yan Yan","doi":"10.1109/IAEAC.2018.8577878","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577878","url":null,"abstract":"Coordinated Multi Point(CoMP) can be employed to reduce interference and improve network throughput in Ultra Dense Network(UDN). However, backhaul capacity is a bottleneck in achieving the goal. Fortunately, proactive caching is a promising technology to alleviate backhaul load. In a cache-enabled system with CoMP, user association is an important problem which can give prominence to its advantage. In this paper, to pursue large network throughput gain at less cost of backhaul load, we formulate user association problem as an Integer Programming(IP) problem. Given the strategy of content placement, we propose an N-Best Binary Particle Swarm Optimization(NB-BPSO) algorithm to solve the problem. Simulation results show that our algorithm improves network throughput and reduces backhaul traffic significantly compared with traditional user association schemes.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"206 1","pages":"964-968"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80431110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577253
Tao Wang, Li-qun Shang, Xianmin Ma
Based on cloud model and matter element theory, and combining the uncertain reasoning characteristics of the cloud model and qualitative and quantitative analysis can be carried out at the same time by matter element theory, a power transformer fault diagnosis method is proposed, which effectively solves the problem of fewer data samples, especially fewer fault data samples. Taking the actual data as an example, the improved matter-element theory model and correlation calculation data are compared, and the results show that the improved matter element theory model has higher diagnostic accuracy than traditional methods. Example analysis verifies the correctness and effectiveness of the method.
{"title":"Application of Cloud Model and Matter Element Theory in Transformer Fault Diagnosis","authors":"Tao Wang, Li-qun Shang, Xianmin Ma","doi":"10.1109/IAEAC.2018.8577253","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577253","url":null,"abstract":"Based on cloud model and matter element theory, and combining the uncertain reasoning characteristics of the cloud model and qualitative and quantitative analysis can be carried out at the same time by matter element theory, a power transformer fault diagnosis method is proposed, which effectively solves the problem of fewer data samples, especially fewer fault data samples. Taking the actual data as an example, the improved matter-element theory model and correlation calculation data are compared, and the results show that the improved matter element theory model has higher diagnostic accuracy than traditional methods. Example analysis verifies the correctness and effectiveness of the method.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"32 1","pages":"2089-2092"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76934725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577926
Xiaosheng Wang, Yongfei Han, Yanchun Yang
Embedded terminal devices are widely used in security-critical systems such as financial payments, and the security of their confidential data is becoming a concern for people. This paper presents a design method of embedded secure dual-core processor for smart terminal devices. The method uses a combination of physical security design and logical security design to protect the confidential data. In the physical security design, a physical isolation between the two cores is implemented through the interconnection network and the dual-core communication module, so that the confidential data runs in a closed environment completely isolated from the actual application to prevent the possibility of the confidential data being drawn at the application development terminal. In the logical security design, using the advantages of dual core, a combination of software and hardware is adopted to implement two-core cooperative encryption to prevent side-channel attacks from leaking confidential data. The proposed processor has very little hardware overhead and performance penalty.
{"title":"A Secure Dual-Core Processor Design for Embedded Terminal Device","authors":"Xiaosheng Wang, Yongfei Han, Yanchun Yang","doi":"10.1109/IAEAC.2018.8577926","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577926","url":null,"abstract":"Embedded terminal devices are widely used in security-critical systems such as financial payments, and the security of their confidential data is becoming a concern for people. This paper presents a design method of embedded secure dual-core processor for smart terminal devices. The method uses a combination of physical security design and logical security design to protect the confidential data. In the physical security design, a physical isolation between the two cores is implemented through the interconnection network and the dual-core communication module, so that the confidential data runs in a closed environment completely isolated from the actual application to prevent the possibility of the confidential data being drawn at the application development terminal. In the logical security design, using the advantages of dual core, a combination of software and hardware is adopted to implement two-core cooperative encryption to prevent side-channel attacks from leaking confidential data. The proposed processor has very little hardware overhead and performance penalty.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"25 1","pages":"1771-1775"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77172816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577842
Hengjie Li, L. Wen, Wei Chen, Xu Gong, Xianqiang Zeng
In order to solve the problem of the rapid and coordinated charging of private electric vehicles, two coordinated charging strategies were formulated under the rapid charging mode. One is to establish a coordinated charging strategy that minimizes the total charging time of electric vehicles, the minimum of power standard deviation and the peak-valley difference of distribution networks. The multi-objective function is normalized to a single-target solution by weighted summation method. The other is to use the total charging time of electric vehicles as a constraint to achieve peak clipping and filling of the distribution network. Taking the collected traffic data of weekdays and holidays in Lanzhou City as an example, two kinds of coordinated optimization schemes were simulated and analyzed on weekdays and holidays conditions respectively, and the genetic algorithm was used to solve the problem. The results show that the total charging time is the smallest when charging is uncoordinated charging, and peak clipping and filling effect of the distribution network is the worst; when electric vehicles do not consider the time target for coordinated charging, peak clipping and filling effect of the distribution network is the best and the total charging time is the longest; when electric vehicles contain the time target for coordinated charging, peak clipping and filling effect of the distribution network is intermediate.
{"title":"The Rapid and Coordinated Charging Strategy for Private Electric Vehicles","authors":"Hengjie Li, L. Wen, Wei Chen, Xu Gong, Xianqiang Zeng","doi":"10.1109/IAEAC.2018.8577842","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577842","url":null,"abstract":"In order to solve the problem of the rapid and coordinated charging of private electric vehicles, two coordinated charging strategies were formulated under the rapid charging mode. One is to establish a coordinated charging strategy that minimizes the total charging time of electric vehicles, the minimum of power standard deviation and the peak-valley difference of distribution networks. The multi-objective function is normalized to a single-target solution by weighted summation method. The other is to use the total charging time of electric vehicles as a constraint to achieve peak clipping and filling of the distribution network. Taking the collected traffic data of weekdays and holidays in Lanzhou City as an example, two kinds of coordinated optimization schemes were simulated and analyzed on weekdays and holidays conditions respectively, and the genetic algorithm was used to solve the problem. The results show that the total charging time is the smallest when charging is uncoordinated charging, and peak clipping and filling effect of the distribution network is the worst; when electric vehicles do not consider the time target for coordinated charging, peak clipping and filling effect of the distribution network is the best and the total charging time is the longest; when electric vehicles contain the time target for coordinated charging, peak clipping and filling effect of the distribution network is intermediate.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"44 1","pages":"196-200"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81436357","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577247
Hui Jiang
on the basis of analyzing and explaining the concept of workflow, this paper analyzes the open source workflow engine jBPM (Java Business Process Management) from the perspectives of architecture, database table structure, authority management and so on. Combining the financial approval process of the enterprise, a prototype of financial approval management system is developed, and the main design ideas and key technologies to be realized are expounded in details. The trial application of the system shows that the jBPM workflow engine has obvious advantages in such aspects as business process customization, automatic management process control, scalability and security. It can meet flexible and changeable needs of the business process, and does improve the efficiency of financial approval, which has guiding significance in engineering practices.
本文在分析和阐述工作流概念的基础上，从体系结构、数据库表结构、权限管理等方面对开源工作流引擎jBPM (Java Business Process Management)进行了分析。结合企业财务审批流程，开发了财务审批管理系统原型，详细阐述了主要设计思想和要实现的关键技术。系统的试用表明，jBPM工作流引擎在业务流程定制、自动化管理流程控制、可扩展性和安全性等方面具有明显的优势。能够满足业务流程灵活多变的需求，提高财务审批效率，在工程实践中具有指导意义。
{"title":"Research and Implementation of Financial Approval System Based on jBPM Engine","authors":"Hui Jiang","doi":"10.1109/IAEAC.2018.8577247","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577247","url":null,"abstract":"on the basis of analyzing and explaining the concept of workflow, this paper analyzes the open source workflow engine jBPM (Java Business Process Management) from the perspectives of architecture, database table structure, authority management and so on. Combining the financial approval process of the enterprise, a prototype of financial approval management system is developed, and the main design ideas and key technologies to be realized are expounded in details. The trial application of the system shows that the jBPM workflow engine has obvious advantages in such aspects as business process customization, automatic management process control, scalability and security. It can meet flexible and changeable needs of the business process, and does improve the efficiency of financial approval, which has guiding significance in engineering practices.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"20 1","pages":"394-399"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81484697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577673
W. Liu, Huaijun Zhou
Boyer et al put forward the first semi-quantum key distribution (BKM2007), which is completely robust while one side is limited. However, its efficiency is not satisfactory. In this paper, we present a high-efficient semi-quantum key distribution. The efficiency for qubits is approximately 50% while that is approximately 25% in the BKM2007 protocol, which means that the sifted key proportion of the SQKD protocol is double as that in the BKM2007 protocol. Furthermore, the proportion of information bits in our SQKD protocol is not lower than that of the BKM2007 protocol. Analysis shows that our protocol is still robust.
{"title":"A new semi-quantum key distribution protocol with high efficiency","authors":"W. Liu, Huaijun Zhou","doi":"10.1109/IAEAC.2018.8577673","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577673","url":null,"abstract":"Boyer et al put forward the first semi-quantum key distribution (BKM2007), which is completely robust while one side is limited. However, its efficiency is not satisfactory. In this paper, we present a high-efficient semi-quantum key distribution. The efficiency for qubits is approximately 50% while that is approximately 25% in the BKM2007 protocol, which means that the sifted key proportion of the SQKD protocol is double as that in the BKM2007 protocol. Furthermore, the proportion of information bits in our SQKD protocol is not lower than that of the BKM2007 protocol. Analysis shows that our protocol is still robust.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"18 1","pages":"2424-2427"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"82490178","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577911
Zepeng Wen, Gongliang Li, Guang-Hong Yang
With the rapid development of internet technology, internet has brought great conveniences to the vast majority of users. However, the exploding of concurrent visits also raised huge challenges to the working capacity of network servers. Thus server cluster and load balancing technologies become a powerful approach to deal with the bottleneck of server terminals. The Nginx-based load balancing method, featured with its high cost performance and extensibility, has been widely applied. This paper raised a Nginx-based dynamic feedback load balancing algorithm. This loading algorithm module is realized by designing an algorithm on the basis of queuing theory and through Nginx plug-in mechanism. Then the paper established an experiment environment, and used Loadrunner software to conduct a comparison test on the performances of the Nginx built-in IP Hash algorithm & Weighted Round-Robin Scheduling and the dynamic feedback algorithm raised in this paper. The results showed that dynamic algorithm raised in this paper can better realize the load balancing.
{"title":"Research and Realization of Nginx-based Dynamic Feedback Load Balancing Algorithm","authors":"Zepeng Wen, Gongliang Li, Guang-Hong Yang","doi":"10.1109/IAEAC.2018.8577911","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577911","url":null,"abstract":"With the rapid development of internet technology, internet has brought great conveniences to the vast majority of users. However, the exploding of concurrent visits also raised huge challenges to the working capacity of network servers. Thus server cluster and load balancing technologies become a powerful approach to deal with the bottleneck of server terminals. The Nginx-based load balancing method, featured with its high cost performance and extensibility, has been widely applied. This paper raised a Nginx-based dynamic feedback load balancing algorithm. This loading algorithm module is realized by designing an algorithm on the basis of queuing theory and through Nginx plug-in mechanism. Then the paper established an experiment environment, and used Loadrunner software to conduct a comparison test on the performances of the Nginx built-in IP Hash algorithm & Weighted Round-Robin Scheduling and the dynamic feedback algorithm raised in this paper. The results showed that dynamic algorithm raised in this paper can better realize the load balancing.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"29 1","pages":"2541-2546"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76326476","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577591
Qingyu Wang, Dongfei Liu
With the development and application of computer technology, cloud computing as a new computing model has become a hot topic in academic research. How to effectively use cluster resources and improve the overall performance of the system have become hot topics and problems in the field of cloud computing. And a reasonable method of load balance is an effective way to solve this problem. Based on the deep study of Shuffled Frog Leading Algorithm, this paper proposes a method of load balance based on improved Shuffled Frog Leaping Algorithm (ISFLA) in cloud computing. Finally, a simulation experiment of the load balance based on improved Shuffled Frog Leaping Algorithm was performed on CloudSim, a simulation platform for cloud computing. Through comparing with the experimental results of other intelligent optimization algorithms, it was found that ISFLA has a better load balance effect and it is feasible.
随着计算机技术的发展和应用，云计算作为一种新的计算模式已经成为学术界研究的热点。如何有效地利用集群资源，提高系统的整体性能已经成为云计算领域的热门话题和问题。而合理的负载均衡方法是解决这一问题的有效途径。在深入研究shuffledfrog Leading Algorithm的基础上，提出了一种基于改进的shuffledfrog leapalgorithm (ISFLA)的云计算负载均衡方法。最后，在云计算仿真平台CloudSim上进行了基于改进的shuffledfrog跳跃算法的负载均衡仿真实验。通过与其他智能优化算法的实验结果对比，发现ISFLA具有更好的负载均衡效果，是可行的。
{"title":"Research on Load Balancing Method in Cloud Computing","authors":"Qingyu Wang, Dongfei Liu","doi":"10.1109/IAEAC.2018.8577591","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577591","url":null,"abstract":"With the development and application of computer technology, cloud computing as a new computing model has become a hot topic in academic research. How to effectively use cluster resources and improve the overall performance of the system have become hot topics and problems in the field of cloud computing. And a reasonable method of load balance is an effective way to solve this problem. Based on the deep study of Shuffled Frog Leading Algorithm, this paper proposes a method of load balance based on improved Shuffled Frog Leaping Algorithm (ISFLA) in cloud computing. Finally, a simulation experiment of the load balance based on improved Shuffled Frog Leaping Algorithm was performed on CloudSim, a simulation platform for cloud computing. Through comparing with the experimental results of other intelligent optimization algorithms, it was found that ISFLA has a better load balance effect and it is feasible.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"75 1","pages":"1489-1493"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76080275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2018-10-01DOI: 10.1109/IAEAC.2018.8577519
Zhang Chao, Chen Ying
In order to detect objects more correctly in pictures, an object detection algorithm based on the Online hard example mining with residual network is put forward, which takes Faster R-CNN as a benchmark. The working method of Faster R-CNN based on deep learning is portrayed, and the drawback and improvement ways of the algorithm are analyzed. More specifically, a residual network is used to replace the original ZF or VGG network to obtain more effective deep convolution feature maps. Besides, in order to strengthen the generalization capacity of the learning network model, the network parameters with hard examples are regenerated during training to make the network training more effectively. Finally, results of the experiments on Pascal VOC2007, Pascal VOC2007+VOC2012 and BIT datasets indicate that compared with Faster R-CNN, our method improves detection accuracy by 3.5%, 7.1%, 6.4% severally on the three datasets.
{"title":"Object detection based on Online hard examples mining with residual network","authors":"Zhang Chao, Chen Ying","doi":"10.1109/IAEAC.2018.8577519","DOIUrl":"https://doi.org/10.1109/IAEAC.2018.8577519","url":null,"abstract":"In order to detect objects more correctly in pictures, an object detection algorithm based on the Online hard example mining with residual network is put forward, which takes Faster R-CNN as a benchmark. The working method of Faster R-CNN based on deep learning is portrayed, and the drawback and improvement ways of the algorithm are analyzed. More specifically, a residual network is used to replace the original ZF or VGG network to obtain more effective deep convolution feature maps. Besides, in order to strengthen the generalization capacity of the learning network model, the network parameters with hard examples are regenerated during training to make the network training more effectively. Finally, results of the experiments on Pascal VOC2007, Pascal VOC2007+VOC2012 and BIT datasets indicate that compared with Faster R-CNN, our method improves detection accuracy by 3.5%, 7.1%, 6.4% severally on the three datasets.","PeriodicalId":6573,"journal":{"name":"2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC)","volume":"33 1","pages":"1634-1638"},"PeriodicalIF":0.0,"publicationDate":"2018-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87953065","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：