Pub Date : 2021-11-22DOI: 10.1109/ICESIT53460.2021.9696753
Linyu Wang, Haiyan Jiang, Yibo Jiang
Increasing proportion of centralized wind power integrated into partial areas of China leads to requirement in sharing both energy and reserve among areas under its inherent hierarchical control structure, and the unbalance power introduced by wind power uncertainty lead to requirement of correction from day ahead to intra-day along with the improvement of wind power prediction precision. In order to address these problems, this paper develops an information integration method integrating complicated relations among fuel cost, total thermal power output, reserve capacity, owned reserve and expectations of loading shedding and wind curtailment within this area into three types of time-related relation curves in different time scale. Furthermore, a multi-time scale tie-line energy and reserve allocation model is proposed, which contains two levels in control structure, two time scales in dispatch sequence and multiple areas integrated with wind farms. The efficiency of the proposed method is tested in 9-bus test system and IEEE 118-bus system. The results show that cross-regional control centre is able to allocate both energy and reserve among areas efficiently with the integrated relation curves. The proposed model not only relieves energy and reserve shortage in partial areas but also allocates them to more urgent areas in a high effectivity manner in both day-ahead and intraday time scale.
{"title":"A Multi-time Scale Tie-line Energy and Reserve Allocation Model Considering Wind Power Uncertainty for Multi-area System in Hierarchical Control Structure","authors":"Linyu Wang, Haiyan Jiang, Yibo Jiang","doi":"10.1109/ICESIT53460.2021.9696753","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696753","url":null,"abstract":"Increasing proportion of centralized wind power integrated into partial areas of China leads to requirement in sharing both energy and reserve among areas under its inherent hierarchical control structure, and the unbalance power introduced by wind power uncertainty lead to requirement of correction from day ahead to intra-day along with the improvement of wind power prediction precision. In order to address these problems, this paper develops an information integration method integrating complicated relations among fuel cost, total thermal power output, reserve capacity, owned reserve and expectations of loading shedding and wind curtailment within this area into three types of time-related relation curves in different time scale. Furthermore, a multi-time scale tie-line energy and reserve allocation model is proposed, which contains two levels in control structure, two time scales in dispatch sequence and multiple areas integrated with wind farms. The efficiency of the proposed method is tested in 9-bus test system and IEEE 118-bus system. The results show that cross-regional control centre is able to allocate both energy and reserve among areas efficiently with the integrated relation curves. The proposed model not only relieves energy and reserve shortage in partial areas but also allocates them to more urgent areas in a high effectivity manner in both day-ahead and intraday time scale.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123283155","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}
Using knowledge distillation to compress pre-trained models such as Bert has proven to be highly effective in text classification tasks. However, the overhead of tuning parameters manually still hinders their application in practice. To alleviate the cost of manual tuning of parameters in training tasks, inspired by the inverse decrease of the word frequency of TF-IDF, this paper proposes an adaptive knowledge distillation method (AKD). This core idea of the method is based on the Cosine similarity score which is calculated by the probabilistic outputs similarity measurement in two networks. The higher the score, the closer the student model's understanding of knowledge is to the teacher model, and the lower the degree of imitation of the teacher model. On the contrary, we need to increase the degree to which the student model imitates the teacher model. Interestingly, this method can improve distillation model quality. Experimental results show that the proposed method significantly improves the precision, recall and F1 value of text classification tasks. However, training speed of AKD is slightly slower than baseline models. This study provides new insights into knowledge distillation.
{"title":"An adaptive knowledge distillation algorithm for text classification","authors":"Zuqin Chen, Tingkai Hu, Chao Chen, Jike Ge, Chengzhi Wu, Wenjun Cheng","doi":"10.1109/ICESIT53460.2021.9696948","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696948","url":null,"abstract":"Using knowledge distillation to compress pre-trained models such as Bert has proven to be highly effective in text classification tasks. However, the overhead of tuning parameters manually still hinders their application in practice. To alleviate the cost of manual tuning of parameters in training tasks, inspired by the inverse decrease of the word frequency of TF-IDF, this paper proposes an adaptive knowledge distillation method (AKD). This core idea of the method is based on the Cosine similarity score which is calculated by the probabilistic outputs similarity measurement in two networks. The higher the score, the closer the student model's understanding of knowledge is to the teacher model, and the lower the degree of imitation of the teacher model. On the contrary, we need to increase the degree to which the student model imitates the teacher model. Interestingly, this method can improve distillation model quality. Experimental results show that the proposed method significantly improves the precision, recall and F1 value of text classification tasks. However, training speed of AKD is slightly slower than baseline models. This study provides new insights into knowledge distillation.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129937823","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 : 2021-11-22DOI: 10.1109/ICESIT53460.2021.9696643
Li-Jung Weng
The in-depth implementation of the “One Belt, One Road” has improved the development of the port economy and perfected the the functions of ports in Guangdong. Therefore, accurate forecasting of the port container throughput is essential for port planning and resource coordination. Taking Guangdong port as an example, the article uses ARIMA, GM (1, 1), ES, ES-GM (1, 1) and ES-ARIMA models to simulate and predict port container throughput. The results show that the optimal model for port throughput prediction is ES-GM (1, 1). In the next five months, the average increase in container port throughput was 2.14 wTEU. Finally, based on the forecast results, suggestions are made for the future development of the port.
{"title":"Prediction of Container Throughput in Guangdong Province Based on Different Model","authors":"Li-Jung Weng","doi":"10.1109/ICESIT53460.2021.9696643","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696643","url":null,"abstract":"The in-depth implementation of the “One Belt, One Road” has improved the development of the port economy and perfected the the functions of ports in Guangdong. Therefore, accurate forecasting of the port container throughput is essential for port planning and resource coordination. Taking Guangdong port as an example, the article uses ARIMA, GM (1, 1), ES, ES-GM (1, 1) and ES-ARIMA models to simulate and predict port container throughput. The results show that the optimal model for port throughput prediction is ES-GM (1, 1). In the next five months, the average increase in container port throughput was 2.14 wTEU. Finally, based on the forecast results, suggestions are made for the future development of the port.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129373895","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 : 2021-11-22DOI: 10.1109/ICESIT53460.2021.9696744
Huang-Chih Lin, Wei Liang, Lan Zhu, Zhen Cheng, Y. Zheng
In order to make the switched-inductor Quasi-Z source inverter have the function of energy bidirectional flowing in some special occasions, the diode in the topology is changed to insulation gate bipolar transistor (IGBT). The influence of capacitance and inductance parameters on zero and pole of bidirectional switched-inductor Quasi-Z source inverter is analyzed by using small signal model. The simulation of SVPWM4 based on shoot-through vector insert is finished.
{"title":"Research on Control Strategy of Improved Bidirectional Quasi-Z Source Inverter","authors":"Huang-Chih Lin, Wei Liang, Lan Zhu, Zhen Cheng, Y. Zheng","doi":"10.1109/ICESIT53460.2021.9696744","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696744","url":null,"abstract":"In order to make the switched-inductor Quasi-Z source inverter have the function of energy bidirectional flowing in some special occasions, the diode in the topology is changed to insulation gate bipolar transistor (IGBT). The influence of capacitance and inductance parameters on zero and pole of bidirectional switched-inductor Quasi-Z source inverter is analyzed by using small signal model. The simulation of SVPWM4 based on shoot-through vector insert is finished.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129622061","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 : 2021-11-22DOI: 10.1109/ICESIT53460.2021.9696810
Xiaoliang Zhang, Jun-ming Zhang, Jiangqiao Li, Limin Zhang, Nan Zou
The underwater acoustic location technique takes advantage of the long-distance propagation of underwater sound wave. With the help of transponder array, buoy or base array which placed in the already known position underwater, on the water or on hull, it can measure the propagation delay, phase difference and more of the acoustic signals that emitted by the target and the location of the target is calculated through geometrical principles. For systems operating at higher frequencies, due to the limitation of physical processes, the spacing of array elements is difficult to satisfy the constraint of half-wave spacing, and the traditional anti-phase ambiguity method is invalid. In order to solve this problem, an anti-phase ambiguity algorithm based on maximum a posterior criterion is investigated. After analyzing the feasibility of the algorithm in theory, the engineering implementation method is given. Then we discuss the influence of array on this method, and verify the accuracy of the method by simulation.
{"title":"Research on Anti-Phase Ambiguity Method for High Frame Rate Ultra-Short Baseline Location System","authors":"Xiaoliang Zhang, Jun-ming Zhang, Jiangqiao Li, Limin Zhang, Nan Zou","doi":"10.1109/ICESIT53460.2021.9696810","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696810","url":null,"abstract":"The underwater acoustic location technique takes advantage of the long-distance propagation of underwater sound wave. With the help of transponder array, buoy or base array which placed in the already known position underwater, on the water or on hull, it can measure the propagation delay, phase difference and more of the acoustic signals that emitted by the target and the location of the target is calculated through geometrical principles. For systems operating at higher frequencies, due to the limitation of physical processes, the spacing of array elements is difficult to satisfy the constraint of half-wave spacing, and the traditional anti-phase ambiguity method is invalid. In order to solve this problem, an anti-phase ambiguity algorithm based on maximum a posterior criterion is investigated. After analyzing the feasibility of the algorithm in theory, the engineering implementation method is given. Then we discuss the influence of array on this method, and verify the accuracy of the method by simulation.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130591018","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 : 2021-11-22DOI: 10.1109/ICESIT53460.2021.9696455
Hongcheng Liao, Wenwen Zhu, Benzhu Zhang, Xiang Zhang, Yu Sun, Cending Wang, Jie Li
Aiming at solving the natural gas leakage detection issue, we propose an improved method based on deep residual network with channel-wise thresholds (DRSN-CW) to improve the detection accuracy with GPLA-12 dataset. In the approach, larger and unequal convolution kernel size are designed in all convolution layers to extend the receptive field in the process of extracting fault feature. Moreover, considering that datasets of natural gas pipeline leakage typically contain large amounts of ambient noise, the soft threshold module of DRSN-CW is combined with designed kernel size to reduce the influence of noise on accuracy of gas pipeline leakage detection. Compared with the-state-of-art techniques (e.g., CNN, DRSN-CW and DRSN-CS), experimental results show that our method outperforms the compared methods.
{"title":"Application of Natural Gas Pipeline Leakage Detection Based on Improved DRSN-CW","authors":"Hongcheng Liao, Wenwen Zhu, Benzhu Zhang, Xiang Zhang, Yu Sun, Cending Wang, Jie Li","doi":"10.1109/ICESIT53460.2021.9696455","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696455","url":null,"abstract":"Aiming at solving the natural gas leakage detection issue, we propose an improved method based on deep residual network with channel-wise thresholds (DRSN-CW) to improve the detection accuracy with GPLA-12 dataset. In the approach, larger and unequal convolution kernel size are designed in all convolution layers to extend the receptive field in the process of extracting fault feature. Moreover, considering that datasets of natural gas pipeline leakage typically contain large amounts of ambient noise, the soft threshold module of DRSN-CW is combined with designed kernel size to reduce the influence of noise on accuracy of gas pipeline leakage detection. Compared with the-state-of-art techniques (e.g., CNN, DRSN-CW and DRSN-CS), experimental results show that our method outperforms the compared methods.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130061693","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 : 2021-11-22DOI: 10.1109/ICESIT53460.2021.9696970
Jing Wang, Yonghong Li
This paper takes Xiahe dialect of Tibetan Amdo as the research object, and there is a special linguistic phenomenon in Xiahe dialect-compound consonants. The paper sorted out the compound consonant system of Xiahe dialect and showed the air flow waveform of different compound consonants. According to the analysis of the air flow parameters of different pre consonant pronunciation, it is concluded that the clarity of pre consonant has a certain impact on the pronunciation duration and average air flow speed.
{"title":"Computer intelligent recognition and studied of the pronunciation of Xiahe dialect database system through speech aerodynamics","authors":"Jing Wang, Yonghong Li","doi":"10.1109/ICESIT53460.2021.9696970","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696970","url":null,"abstract":"This paper takes Xiahe dialect of Tibetan Amdo as the research object, and there is a special linguistic phenomenon in Xiahe dialect-compound consonants. The paper sorted out the compound consonant system of Xiahe dialect and showed the air flow waveform of different compound consonants. According to the analysis of the air flow parameters of different pre consonant pronunciation, it is concluded that the clarity of pre consonant has a certain impact on the pronunciation duration and average air flow speed.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130169109","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 : 2021-11-22DOI: 10.1109/ICESIT53460.2021.9696701
P. Yao
Some studies show that the closure and reopening orders brought by covid-19 have had a negative impact on the residential real estate market. Generally speaking, real estate sales decreased significantly during this period, such as office buildings, shopping centers and family houses. Although the overall situation is declining, there are also some new situations. For example, people's desire for spacious family space caused by home office leads to an increase in the demand for large houses in the suburbs. This paper mainly compares the sales differences between suburban family houses and urban family houses in San Francisco and New York in the real estate market during covid-19. The data come from multiple dimensions such as house listing price on the real estate sales website, Machine learning methods could be used for analysis. This paper proposed a multi-modal joint attention seq2seq method to analyze these differences and the reasons for the differences. The experimental results show that one of the possible reasons the house price change in San Francisco is that there are more high-tech job position and their family income is higher than the average level of other regions.
{"title":"A Multi-modal Attention-based Seq2eq Model for Predicting Real-estate Prices","authors":"P. Yao","doi":"10.1109/ICESIT53460.2021.9696701","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696701","url":null,"abstract":"Some studies show that the closure and reopening orders brought by covid-19 have had a negative impact on the residential real estate market. Generally speaking, real estate sales decreased significantly during this period, such as office buildings, shopping centers and family houses. Although the overall situation is declining, there are also some new situations. For example, people's desire for spacious family space caused by home office leads to an increase in the demand for large houses in the suburbs. This paper mainly compares the sales differences between suburban family houses and urban family houses in San Francisco and New York in the real estate market during covid-19. The data come from multiple dimensions such as house listing price on the real estate sales website, Machine learning methods could be used for analysis. This paper proposed a multi-modal joint attention seq2seq method to analyze these differences and the reasons for the differences. The experimental results show that one of the possible reasons the house price change in San Francisco is that there are more high-tech job position and their family income is higher than the average level of other regions.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128871931","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 : 2021-11-22DOI: 10.1109/ICESIT53460.2021.9696872
Bin Zhao, Boyu Zhao, Pengfei Li
With the development of society, the application of abnormal behavior detection in the field of public safety has become more and more extensive. We propose a frame prediction video behavior anomaly detection model based on Generative Adversarial Network (GAN). We use the U-net network with the feature storage module and variance attention mechanism as the generator, which not only increases the network's sensitivity to the movement part of the sample, but also reduces the network's learning ability and limits the network's ability to predict abnormal samples. For the discriminant model, we have added a channel and spatial attention mechanism to the Markov discriminator to improve the discrimination ability, which is conducive to improving the quality of future frame generation. Compared with the existing abnormal behavior detection methods, our proposed model achieves excellent detection performance.
{"title":"Video Anomaly Detection Based on Frame Prediction of Generative Adversarial Network","authors":"Bin Zhao, Boyu Zhao, Pengfei Li","doi":"10.1109/ICESIT53460.2021.9696872","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696872","url":null,"abstract":"With the development of society, the application of abnormal behavior detection in the field of public safety has become more and more extensive. We propose a frame prediction video behavior anomaly detection model based on Generative Adversarial Network (GAN). We use the U-net network with the feature storage module and variance attention mechanism as the generator, which not only increases the network's sensitivity to the movement part of the sample, but also reduces the network's learning ability and limits the network's ability to predict abnormal samples. For the discriminant model, we have added a channel and spatial attention mechanism to the Markov discriminator to improve the discrimination ability, which is conducive to improving the quality of future frame generation. Compared with the existing abnormal behavior detection methods, our proposed model achieves excellent detection performance.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122336965","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}
For the cross-border tracking needs of target persons in real complex scenes, a person re-identification algorithm based on a multi-module convolution neural network is proposed to solve the problem of person search and matching caused by person scale change, light change, posture change and other factors in the real environment. The algorithm takes ResNet50 as the backbone network of feature extraction. The STN network module is embedded into the backbone network to overcome the impact of person scale change. The IBN network module is integrated for person image color correction to compensate for the influence of illumination change in the real scene. And A person multi-branch feature extraction module is designed to effectively reduce the impact of person posture changes. Through person image feature expression and measurement learning calculation, person similarity matching across cameras is realized. Experiments show that this method has good performance in real complex scene test data, and its Rank-1 and mAP are 98.30% and 95.78% respectively. It can be used for person matching and search in a real complex environment, and has certain practical value.
{"title":"Person re-identification algorithm based on multi-module convolutional neural network","authors":"Huan Lei, Zeyu Jiao, Junhao Lin, Zaili Chen, Chentong Li, Z. Zhong","doi":"10.1109/ICESIT53460.2021.9696542","DOIUrl":"https://doi.org/10.1109/ICESIT53460.2021.9696542","url":null,"abstract":"For the cross-border tracking needs of target persons in real complex scenes, a person re-identification algorithm based on a multi-module convolution neural network is proposed to solve the problem of person search and matching caused by person scale change, light change, posture change and other factors in the real environment. The algorithm takes ResNet50 as the backbone network of feature extraction. The STN network module is embedded into the backbone network to overcome the impact of person scale change. The IBN network module is integrated for person image color correction to compensate for the influence of illumination change in the real scene. And A person multi-branch feature extraction module is designed to effectively reduce the impact of person posture changes. Through person image feature expression and measurement learning calculation, person similarity matching across cameras is realized. Experiments show that this method has good performance in real complex scene test data, and its Rank-1 and mAP are 98.30% and 95.78% respectively. It can be used for person matching and search in a real complex environment, and has certain practical value.","PeriodicalId":164745,"journal":{"name":"2021 IEEE International Conference on Emergency Science and Information Technology (ICESIT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116310655","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：