Pub Date : 2020-11-20DOI: 10.1109/AUTEEE50969.2020.9315607
Dongfang Hu, Bin Xu, Jun Wang, Linfeng Han, Jiayi Liu
In this paper, we propose a malware detection method based on feature library and machine learning. By using a combination of static and dynamic feature extraction method, we select 8 types of static features to build a feature library. In addition, for potentially unknown malwares, we use 9 groups of dynamic features to train a support vector machine model, and give interpretable detection results based on the influence of different features. To verify the performance of our method, we conducted various experiments on a total of 129,013 malware samples and compared the results with other schemes, demonstrating the effectiveness of our method.
{"title":"Malware Detection Based on Feature Library and Machine Learning","authors":"Dongfang Hu, Bin Xu, Jun Wang, Linfeng Han, Jiayi Liu","doi":"10.1109/AUTEEE50969.2020.9315607","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315607","url":null,"abstract":"In this paper, we propose a malware detection method based on feature library and machine learning. By using a combination of static and dynamic feature extraction method, we select 8 types of static features to build a feature library. In addition, for potentially unknown malwares, we use 9 groups of dynamic features to train a support vector machine model, and give interpretable detection results based on the influence of different features. To verify the performance of our method, we conducted various experiments on a total of 129,013 malware samples and compared the results with other schemes, demonstrating the effectiveness of our method.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"5 1","pages":"205-213"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72965306","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}
With the continuous development of power systems, frequency control faces new challenges, and the demand side should play an increasingly important role in frequency regulation. In this paper, the callback characteristic of demand side resource is proposed firstly. The characteristic is caused by the comfort requirements of the users. Choosing air conditioner as an example, after participating in frequency control for some period, its load should be withdrawn to maintain acceptable room temperature. Meanwhile, above callback characteristic will deteriorate the frequency control performance, and thus a new frequency control strategy is proposed by coordinating the resources of generation side and demand side. The simulation results show that the strategy has better economic performance compared with the traditional precision load shedding method.
{"title":"Coordinated Frequency Control of Generation and Demand Side Considering Demand Side Resource Callback Characteristic","authors":"Zhenxing Li, Xin Shan, Zhixian Wang, Kun Yuan, Ying Wang, Kaifeng Zhang","doi":"10.1109/AUTEEE50969.2020.9315685","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315685","url":null,"abstract":"With the continuous development of power systems, frequency control faces new challenges, and the demand side should play an increasingly important role in frequency regulation. In this paper, the callback characteristic of demand side resource is proposed firstly. The characteristic is caused by the comfort requirements of the users. Choosing air conditioner as an example, after participating in frequency control for some period, its load should be withdrawn to maintain acceptable room temperature. Meanwhile, above callback characteristic will deteriorate the frequency control performance, and thus a new frequency control strategy is proposed by coordinating the resources of generation side and demand side. The simulation results show that the strategy has better economic performance compared with the traditional precision load shedding method.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"115 1","pages":"248-252"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75751359","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 : 2020-11-20DOI: 10.1109/AUTEEE50969.2020.9315592
Haoqin Ji, T. Wan, Wanan Xiong, Jingyi Liao
Specific emitter identification (SEI) is an association of radar signal to specific emitter primarily. SEI has been widely used in military and civilian spectrum management applications. We propose a SEI method based on deep learning (convolutional neural network), which uses the characteristics of the received steady-state signal. Particularly, we calculate the bispectrum of the signal as the unique feature. Then, we use surrounding-line bispectrum to reduce the influence of noise. Finally, CNN is used to identify specific emitters by using the surrounding-line bispectrum. This method basically extracts the overall feature information hidden in the original signal. This can be used to improve recognition performance. The simulation results verify our conclusion that the proposed method is better than other existing solutions in the literature.
{"title":"A method for specific emitter identification based on surrounding-line bispectrum and convolutional neural network","authors":"Haoqin Ji, T. Wan, Wanan Xiong, Jingyi Liao","doi":"10.1109/AUTEEE50969.2020.9315592","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315592","url":null,"abstract":"Specific emitter identification (SEI) is an association of radar signal to specific emitter primarily. SEI has been widely used in military and civilian spectrum management applications. We propose a SEI method based on deep learning (convolutional neural network), which uses the characteristics of the received steady-state signal. Particularly, we calculate the bispectrum of the signal as the unique feature. Then, we use surrounding-line bispectrum to reduce the influence of noise. Finally, CNN is used to identify specific emitters by using the surrounding-line bispectrum. This method basically extracts the overall feature information hidden in the original signal. This can be used to improve recognition performance. The simulation results verify our conclusion that the proposed method is better than other existing solutions in the literature.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"86 1","pages":"328-332"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74835531","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 : 2020-11-20DOI: 10.1109/AUTEEE50969.2020.9315668
Gen Li, Qiusheng Zheng, Long Zhang, Suzhou Guo, Liyue Niu
As an important task of natural language processing, chinese text sentiment analysis aims to analyze the comprehensive sentiment polarity of chinese text. With the emergence of various deep neural network models, sentiment analysis tasks have once again made significant progress. However, these neural network models could not accurately capture sentiment information on sentiment analysis tasks, which leads to their instability. In order to enable the model to explicitly learn the sentiment knowledge in chinese text, this paper proposes a sentiment information based network model(SINM). We use Transfomer encoder and LSTM as model components. With the help of Chinese emotional dictionary, we can automatically find sentiment knowledge in chinese text. In SINM, we designed a hybrid task learning method to learn valuable emotional expressions and predict sentiment tendencies. First of all, SINM needs to learn the sentiment knowledge in the text. Under the auxiliary influence of emotional information, SINM will pay more attention to sentiment information rather than useless information. Experiments on the dataset of ChnSentiCorp and ChnFoodReviews have found that SINM can achieve better performance and generalization ability than most existing methods.
{"title":"Sentiment Infomation based Model For Chinese text Sentiment Analysis","authors":"Gen Li, Qiusheng Zheng, Long Zhang, Suzhou Guo, Liyue Niu","doi":"10.1109/AUTEEE50969.2020.9315668","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315668","url":null,"abstract":"As an important task of natural language processing, chinese text sentiment analysis aims to analyze the comprehensive sentiment polarity of chinese text. With the emergence of various deep neural network models, sentiment analysis tasks have once again made significant progress. However, these neural network models could not accurately capture sentiment information on sentiment analysis tasks, which leads to their instability. In order to enable the model to explicitly learn the sentiment knowledge in chinese text, this paper proposes a sentiment information based network model(SINM). We use Transfomer encoder and LSTM as model components. With the help of Chinese emotional dictionary, we can automatically find sentiment knowledge in chinese text. In SINM, we designed a hybrid task learning method to learn valuable emotional expressions and predict sentiment tendencies. First of all, SINM needs to learn the sentiment knowledge in the text. Under the auxiliary influence of emotional information, SINM will pay more attention to sentiment information rather than useless information. Experiments on the dataset of ChnSentiCorp and ChnFoodReviews have found that SINM can achieve better performance and generalization ability than most existing methods.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"34 1","pages":"366-371"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79764894","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 : 2020-11-20DOI: 10.1109/AUTEEE50969.2020.9315669
Yuechao Wu, Liwei Liu, Yuanlin Luo, Bo Zheng, Zhenqiu Feng, Dongdong Zhang, Xinyu Wang
The smart hydropower station (SHS) is an important research direction of the power system. The smart equipment is the most important part of the SHS. The holographic management of equipment is the foundation of the smart equipment. In order to meet the requirements of the SHS and smart equipment, a holographic management system for equipment has been developed with the construction of hydropower stations. During the design period, the 3D digital models have been built by 3D design software and assigned with KKS codes. During operation and maintenance period, firstly, the images of hydropower stations have been collected as well as posted onto surfaces of models. Subsequently, the 3D hydropower station has been assembled, the lightweight version has been released. And then, the data have been integrated and the relationships between models and data have been established based on KKS codes. Finally, the holographic management system has been developed.
{"title":"The Holographic Management System Based on KKS Code and 3D Digital Model for Equipment in Smart Hydropower Stations","authors":"Yuechao Wu, Liwei Liu, Yuanlin Luo, Bo Zheng, Zhenqiu Feng, Dongdong Zhang, Xinyu Wang","doi":"10.1109/AUTEEE50969.2020.9315669","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315669","url":null,"abstract":"The smart hydropower station (SHS) is an important research direction of the power system. The smart equipment is the most important part of the SHS. The holographic management of equipment is the foundation of the smart equipment. In order to meet the requirements of the SHS and smart equipment, a holographic management system for equipment has been developed with the construction of hydropower stations. During the design period, the 3D digital models have been built by 3D design software and assigned with KKS codes. During operation and maintenance period, firstly, the images of hydropower stations have been collected as well as posted onto surfaces of models. Subsequently, the 3D hydropower station has been assembled, the lightweight version has been released. And then, the data have been integrated and the relationships between models and data have been established based on KKS codes. Finally, the holographic management system has been developed.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"23 1","pages":"259-263"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84157522","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 : 2020-11-20DOI: 10.1109/AUTEEE50969.2020.9315723
Yin Tian, Chenchen Wei, Dongwei Xu
Accurate prediction traffic flow is one of the most critical works of the intelligent transport system (ITS). Accurate prediction results can provide better conditions for traffic guidance, management, and control. However, many existing traffic flow prediction methods are not particularly satisfactory in practical applications. In this paper, the stack auto-encoder (SAE) and long short-term memory network (LSTM) are combined for traffic flow prediction, in which SAE is used to obtain spatial features, while LSTM extracts temporal features of traffic flow. Then, the features from SAE and LSTM are combined to predict the traffic flow state. The real-time traffic flow data from Beijing is used to evaluate the performance of the proposed method. Experimental results show that the performance of the proposed method is better than some well-known prediction models.
{"title":"Traffic Flow Prediction Based on Stack AutoEncoder and Long Short-Term Memory Network","authors":"Yin Tian, Chenchen Wei, Dongwei Xu","doi":"10.1109/AUTEEE50969.2020.9315723","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315723","url":null,"abstract":"Accurate prediction traffic flow is one of the most critical works of the intelligent transport system (ITS). Accurate prediction results can provide better conditions for traffic guidance, management, and control. However, many existing traffic flow prediction methods are not particularly satisfactory in practical applications. In this paper, the stack auto-encoder (SAE) and long short-term memory network (LSTM) are combined for traffic flow prediction, in which SAE is used to obtain spatial features, while LSTM extracts temporal features of traffic flow. Then, the features from SAE and LSTM are combined to predict the traffic flow state. The real-time traffic flow data from Beijing is used to evaluate the performance of the proposed method. Experimental results show that the performance of the proposed method is better than some well-known prediction models.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"50 1","pages":"385-388"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72891245","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 : 2020-11-20DOI: 10.1109/AUTEEE50969.2020.9315690
S. Lv, Haizheng Zhang, Feihu Bao
For the safe usage of missile fuel tank in a long-term preservation process, the health status of the storage tank must be evaluated and managed accurately. Currently, the health management strategy has gradually evolved from Time-based maintenance (TBM) to Preventive maintenance (PM). With artificial intelligence (AI) applied to process the big data, the strategy of tank storage health management is now able to make precis predictions and guidance. The basic data is acquired from various databases, and the prediction of the structural performance of the storage tank system is accomplished by a series of simulation models intelligently. The modules include data fused long storage evaluation module, corrosion depth prediction module, elastic modulus drop prediction module, and creep damage analysis module. With on-site monitoring of data, a decision tree model based on artificial intelligence is constructed to provide decision support for the use of the missile propellant tank, leading to a more effective, time-saving, and accurate control strategy.
{"title":"Health Management Control Strategy of Tank Storage Based on Artificial Intelligence","authors":"S. Lv, Haizheng Zhang, Feihu Bao","doi":"10.1109/AUTEEE50969.2020.9315690","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315690","url":null,"abstract":"For the safe usage of missile fuel tank in a long-term preservation process, the health status of the storage tank must be evaluated and managed accurately. Currently, the health management strategy has gradually evolved from Time-based maintenance (TBM) to Preventive maintenance (PM). With artificial intelligence (AI) applied to process the big data, the strategy of tank storage health management is now able to make precis predictions and guidance. The basic data is acquired from various databases, and the prediction of the structural performance of the storage tank system is accomplished by a series of simulation models intelligently. The modules include data fused long storage evaluation module, corrosion depth prediction module, elastic modulus drop prediction module, and creep damage analysis module. With on-site monitoring of data, a decision tree model based on artificial intelligence is constructed to provide decision support for the use of the missile propellant tank, leading to a more effective, time-saving, and accurate control strategy.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"31 1","pages":"91-95"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78869851","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 : 2020-11-20DOI: 10.1109/AUTEEE50969.2020.9315698
Jinyong Yin, Jian Yang, Hongbin Yang
Virtual machine has become the foundation of cloud computing with its flexibility and manageability in resource allocation. But based on existing virtualization technologies, a virtual machine can only utilize computing resources on one physical host, which results in fewer computing resources and low computing performance and cannot meet the high-performance computing needs such as artificial intelligence. In this paper, a PCIe cluster based device recompose and resource sharing algorithm is proposed, which enables virtual machines to use the entire cluster’s computing devices in need via PCIe bus. Experimental results show that the algorithm greatly improves the computing resources of virtual machines, improves the computing performance of virtual machines, and meets the high-performance computing needs.
{"title":"An IO device recompose algorithm supporting virtual machine migration on multiple hosts","authors":"Jinyong Yin, Jian Yang, Hongbin Yang","doi":"10.1109/AUTEEE50969.2020.9315698","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315698","url":null,"abstract":"Virtual machine has become the foundation of cloud computing with its flexibility and manageability in resource allocation. But based on existing virtualization technologies, a virtual machine can only utilize computing resources on one physical host, which results in fewer computing resources and low computing performance and cannot meet the high-performance computing needs such as artificial intelligence. In this paper, a PCIe cluster based device recompose and resource sharing algorithm is proposed, which enables virtual machines to use the entire cluster’s computing devices in need via PCIe bus. Experimental results show that the algorithm greatly improves the computing resources of virtual machines, improves the computing performance of virtual machines, and meets the high-performance computing needs.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"23 1","pages":"355-358"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"81959294","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 : 2020-11-20DOI: 10.1109/AUTEEE50969.2020.9315588
Dongfang Hu, Bin Xu, Jun Wang, Linfeng Han, Jiayi Liu
With the development of the Internet, the amount of information on the Internet is increasing rapidly, which makes it difficult for people to select the information they really want. A recommendation system is an effective way to solve this problem. Fake users can be injected by criminals to attack the recommendation system; therefore, accurate identification of fake users is a necessary feature of the recommendation system. Existing fake user detection algorithms focus on designing recognition methods for different types of attacks and have limited detection capabilities against unknown or hybrid attacks. The use of deep learning models can automate the extraction of false user scoring features, but neural network models are not applicable to discrete user scoring data. In this paper, random walking is used to rearrange the otherwise discrete user rating data into a rating feature matrix with spatial continuity. The rating data and the text data have some similarity in the distribution mode. By effective analogy, the TextCNN model originally used in NLP domain can be improved and applied to the classification task of rating feature matrix. Combining the ideas of random walking and word vector processing, this paper proposes a TextCNN detection model for user rating data. To verify the validity of the proposed model, the model is tested on MoiveLens dataset against 7 different attack detection algorithms, and exhibits better performance when compared with 4 attack detection algorithms. Especially for the Aop attack, the proposed model has nearly 100% detection performance with F1 – value as the evaluation index.
{"title":"A Shilling Attack Model Based on TextCNN","authors":"Dongfang Hu, Bin Xu, Jun Wang, Linfeng Han, Jiayi Liu","doi":"10.1109/AUTEEE50969.2020.9315588","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315588","url":null,"abstract":"With the development of the Internet, the amount of information on the Internet is increasing rapidly, which makes it difficult for people to select the information they really want. A recommendation system is an effective way to solve this problem. Fake users can be injected by criminals to attack the recommendation system; therefore, accurate identification of fake users is a necessary feature of the recommendation system. Existing fake user detection algorithms focus on designing recognition methods for different types of attacks and have limited detection capabilities against unknown or hybrid attacks. The use of deep learning models can automate the extraction of false user scoring features, but neural network models are not applicable to discrete user scoring data. In this paper, random walking is used to rearrange the otherwise discrete user rating data into a rating feature matrix with spatial continuity. The rating data and the text data have some similarity in the distribution mode. By effective analogy, the TextCNN model originally used in NLP domain can be improved and applied to the classification task of rating feature matrix. Combining the ideas of random walking and word vector processing, this paper proposes a TextCNN detection model for user rating data. To verify the validity of the proposed model, the model is tested on MoiveLens dataset against 7 different attack detection algorithms, and exhibits better performance when compared with 4 attack detection algorithms. Especially for the Aop attack, the proposed model has nearly 100% detection performance with F1 – value as the evaluation index.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"42 1","pages":"282-289"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89927083","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 : 2020-11-20DOI: 10.1109/AUTEEE50969.2020.9315683
Haoyang Sun, Peng Wang, Dong Zhang, Cui Ni, Hongbo Zhang
The feature points extracted by the traditional ORB algorithm are not evenly distributed, redundant and have no scale invariance. To solve this problem, this paper improved the traditional ORB algorithm and proposed an optimized feature point extraction method. The image is divided into regions firstly. According to the total number of feature points to be extracted and the number of divided regions, the algorithm calculates the number of feature points to be extracted for each region, which solves the problem of feature point overlap and redundancy in the feature point extraction process. By constructing the image pyramid and extracting feature points on each layer, the problem that the feature points extracted by ORB algorithm do not have scale invariance is solved. The experimental results show that the feature points extracted by our algorithm are more uniform and reasonable without losing the accuracy of image matching, and the extraction speed is about 16% higher than that of the traditional ORB algorithm.
{"title":"An Improved ORB Algorithm Based on Optimized Feature Point Extraction","authors":"Haoyang Sun, Peng Wang, Dong Zhang, Cui Ni, Hongbo Zhang","doi":"10.1109/AUTEEE50969.2020.9315683","DOIUrl":"https://doi.org/10.1109/AUTEEE50969.2020.9315683","url":null,"abstract":"The feature points extracted by the traditional ORB algorithm are not evenly distributed, redundant and have no scale invariance. To solve this problem, this paper improved the traditional ORB algorithm and proposed an optimized feature point extraction method. The image is divided into regions firstly. According to the total number of feature points to be extracted and the number of divided regions, the algorithm calculates the number of feature points to be extracted for each region, which solves the problem of feature point overlap and redundancy in the feature point extraction process. By constructing the image pyramid and extracting feature points on each layer, the problem that the feature points extracted by ORB algorithm do not have scale invariance is solved. The experimental results show that the feature points extracted by our algorithm are more uniform and reasonable without losing the accuracy of image matching, and the extraction speed is about 16% higher than that of the traditional ORB algorithm.","PeriodicalId":6767,"journal":{"name":"2020 IEEE 3rd International Conference on Automation, Electronics and Electrical Engineering (AUTEEE)","volume":"5 1","pages":"389-394"},"PeriodicalIF":0.0,"publicationDate":"2020-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75605998","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：