FPGA was a programmable chip with powerful high parallel computing capabilities. Through the FPAG&ARM collaborative processing of neural networks, it can improve computing efficiency and reduce energy consumption. Aiming at a large number of matrix operation problems involved in the AlexNet network, this paper uses the OPENBLAS acceleration algorithm to optimize the matrix operation, and proposes a research and implementation of an embedded image classification method based on ZYNQ. This paper uses FPGA to realize the real-time image acquisition, takes the acquired image as the input of the convolutional neural network model, uses the AlexNet network on the ARM side to realize the classification of the image, and finally deploys the neural network model to the ZYNQ-7000 platform. The experimental results show that in the case of ensuring that the accuracy is not reduced during the image classification process, compared with using the AlexNet network to achieve image classification in 23.4s, using the OPENBLAS acceleration algorithm to accelerate the matrix calculation in the AlexNet network, the image classification consumes time is about 4.5s, and the classification performance has been improved by nearly 5.2 time.
{"title":"Research and Implementation of an Embedded Image Classification Method Based on ZYNQ","authors":"Jiangbo Wang, Zhenyu Yin, Fulong Xu, Feiqing Zhang, Guangyuan Xu","doi":"10.1109/ICTech55460.2022.00024","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00024","url":null,"abstract":"FPGA was a programmable chip with powerful high parallel computing capabilities. Through the FPAG&ARM collaborative processing of neural networks, it can improve computing efficiency and reduce energy consumption. Aiming at a large number of matrix operation problems involved in the AlexNet network, this paper uses the OPENBLAS acceleration algorithm to optimize the matrix operation, and proposes a research and implementation of an embedded image classification method based on ZYNQ. This paper uses FPGA to realize the real-time image acquisition, takes the acquired image as the input of the convolutional neural network model, uses the AlexNet network on the ARM side to realize the classification of the image, and finally deploys the neural network model to the ZYNQ-7000 platform. The experimental results show that in the case of ensuring that the accuracy is not reduced during the image classification process, compared with using the AlexNet network to achieve image classification in 23.4s, using the OPENBLAS acceleration algorithm to accelerate the matrix calculation in the AlexNet network, the image classification consumes time is about 4.5s, and the classification performance has been improved by nearly 5.2 time.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121219006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1109/ICTech55460.2022.00088
Yiying Zhang, Nan Zhang, Yiyang Liu, Caixia Ma, Delong Wang
Aiming at the problems of low recognition rate and easy to be disturbed by noise in the process of single-mode speech emotion recognition, this paper proposes a speech emotion analysis method based on multi feature fusion of speech and semantics. This method uses opensmile to extract acoustic features and Bi long and short term memory network (Bi-LSTM) to extract semantic features, then carries out feature data fusion, and then inputs the fused data into SVM classification model to obtain the final emotion classification result. This method can effectively solve the shortcomings of single-mode emotion recognition and improve the efficiency and accuracy of recognition.
{"title":"A Multimodal Emotion Recognition Method Based on Speech-Text","authors":"Yiying Zhang, Nan Zhang, Yiyang Liu, Caixia Ma, Delong Wang","doi":"10.1109/ICTech55460.2022.00088","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00088","url":null,"abstract":"Aiming at the problems of low recognition rate and easy to be disturbed by noise in the process of single-mode speech emotion recognition, this paper proposes a speech emotion analysis method based on multi feature fusion of speech and semantics. This method uses opensmile to extract acoustic features and Bi long and short term memory network (Bi-LSTM) to extract semantic features, then carries out feature data fusion, and then inputs the fused data into SVM classification model to obtain the final emotion classification result. This method can effectively solve the shortcomings of single-mode emotion recognition and improve the efficiency and accuracy of recognition.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115712853","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1109/ICTech55460.2022.00077
Chuang Zheng, Zhanguo Wang, Jin He
Compared with the current question answering system, similarity matching methods are largely separated into two categories: deep learning methods and conventional ways. Conventional ways rely heavily on artificial features, have weak generalization ability, and insufficient accuracy. RNN and CNN also have text global feature extraction. Limitations. This paper proposes a BERT-based hybrid question answering matching model, which uses the BERT -base pre-training model to capture and represent the semantic information of the QAS sentence and the semantic relevance between the two. The feature vector generated by the BERT model is used as Bi-LSTM _ GCN for the input of the model, feature extraction is performed to further obtain the syntactic features of the sentence, and finally the attention mechanism is added to find the target answer, and the effectiveness of the proposed algorithm is verified on the two types of data sets.
{"title":"BERT-Based Mixed Question Answering Matching Model","authors":"Chuang Zheng, Zhanguo Wang, Jin He","doi":"10.1109/ICTech55460.2022.00077","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00077","url":null,"abstract":"Compared with the current question answering system, similarity matching methods are largely separated into two categories: deep learning methods and conventional ways. Conventional ways rely heavily on artificial features, have weak generalization ability, and insufficient accuracy. RNN and CNN also have text global feature extraction. Limitations. This paper proposes a BERT-based hybrid question answering matching model, which uses the BERT -base pre-training model to capture and represent the semantic information of the QAS sentence and the semantic relevance between the two. The feature vector generated by the BERT model is used as Bi-LSTM _ GCN for the input of the model, feature extraction is performed to further obtain the syntactic features of the sentence, and finally the attention mechanism is added to find the target answer, and the effectiveness of the proposed algorithm is verified on the two types of data sets.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"14 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127051228","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1109/ICTech55460.2022.00064
Xin Deng
In order to make use of data mining technology to do ideological and political education work well, it is necessary to start with the construction of a systematic online education platform, and the integration of interest mining and learning module function design, which can not only obtain a more perfect platform function, but also improve the effect of online learning experience. Therefore, after understanding the improved Apriori interest mining algorithm based on fuzzy theory, this paper analyzes the implementation effect of the actual platform design according to the actual ideological and political education online platform design.
{"title":"Practice and Research of Ideological and Political Education Based on Data Mining Technology","authors":"Xin Deng","doi":"10.1109/ICTech55460.2022.00064","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00064","url":null,"abstract":"In order to make use of data mining technology to do ideological and political education work well, it is necessary to start with the construction of a systematic online education platform, and the integration of interest mining and learning module function design, which can not only obtain a more perfect platform function, but also improve the effect of online learning experience. Therefore, after understanding the improved Apriori interest mining algorithm based on fuzzy theory, this paper analyzes the implementation effect of the actual platform design according to the actual ideological and political education online platform design.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125311124","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1109/ICTech55460.2022.00087
Xianxin Shao, Xiaojun Xia, Jia-Yin Song
To address the problems of low accuracy and inaccurate classification of surface defects detection that occur on metallic steel, this paper proposes a method to improve the accuracy of surface defect detection by adjusting the network structure of the YOLOv3 algorithm model[1]. First, the k-means++ algorithm is used for clustering to improve the matching of prior frames of different scales and feature layers by increasing the scale difference of prior frames. Secondly, the algorithm improves the recognition rate of small defective targets by adding 104×104 feature layers. Finally, the spatial pyramid pooling module is added to improve the recognition accuracy of target features after extracting feature layers of different scales from the backbone feature network. The experimental results show that the improved YOLOv3 algorithm model achieves an average accuracy of 76% on the test set and is 9% better than the original YOLOv3 algorithm than Faster-R-CNN1 in terms of detection performance.
{"title":"Design and Research of Metal Surface Defect Detection Based on Machine Vision","authors":"Xianxin Shao, Xiaojun Xia, Jia-Yin Song","doi":"10.1109/ICTech55460.2022.00087","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00087","url":null,"abstract":"To address the problems of low accuracy and inaccurate classification of surface defects detection that occur on metallic steel, this paper proposes a method to improve the accuracy of surface defect detection by adjusting the network structure of the YOLOv3 algorithm model[1]. First, the k-means++ algorithm is used for clustering to improve the matching of prior frames of different scales and feature layers by increasing the scale difference of prior frames. Secondly, the algorithm improves the recognition rate of small defective targets by adding 104×104 feature layers. Finally, the spatial pyramid pooling module is added to improve the recognition accuracy of target features after extracting feature layers of different scales from the backbone feature network. The experimental results show that the improved YOLOv3 algorithm model achieves an average accuracy of 76% on the test set and is 9% better than the original YOLOv3 algorithm than Faster-R-CNN1 in terms of detection performance.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122938237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1109/ICTech55460.2022.00109
Hongliang Wang, XinXin Lu
In order to improve the semantic understanding of the input question in the question answering system, a question answering system based on knowledge representation is constructed, which is composed of named entity recognition and question matching. The named entity recognition method based on Bert+BiLSTM+CRF is used, and the BGCNN model proposed in this paper is used for question matching. BGCNN is a model combining Bert, neural network and Siamese network. The average F1 value of the system on the financial data set is 0.9007, which is not a small improvement compared with the previous model.
{"title":"Question Answering System with Enhancing Sentence Embedding","authors":"Hongliang Wang, XinXin Lu","doi":"10.1109/ICTech55460.2022.00109","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00109","url":null,"abstract":"In order to improve the semantic understanding of the input question in the question answering system, a question answering system based on knowledge representation is constructed, which is composed of named entity recognition and question matching. The named entity recognition method based on Bert+BiLSTM+CRF is used, and the BGCNN model proposed in this paper is used for question matching. BGCNN is a model combining Bert, neural network and Siamese network. The average F1 value of the system on the financial data set is 0.9007, which is not a small improvement compared with the previous model.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114702261","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1109/ICTech55460.2022.00065
Lai Ding
The College English teaching quality analysis system based on cloud computing framework needs to use data mining algorithm to implement optimization processing, which can not only fully grasp the practical English teaching efficiency, but also find out the factors that affect students’ learning ability. Therefore, based on the understanding of data mining and the construction of college English teaching quality assessment system, this paper aims at the application status of data mining in college English teaching quality analysis in recent years, and evaluates the application effect by combining with experimental design.
{"title":"Data Mining in the University English Teaching Quality Analysis and Research","authors":"Lai Ding","doi":"10.1109/ICTech55460.2022.00065","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00065","url":null,"abstract":"The College English teaching quality analysis system based on cloud computing framework needs to use data mining algorithm to implement optimization processing, which can not only fully grasp the practical English teaching efficiency, but also find out the factors that affect students’ learning ability. Therefore, based on the understanding of data mining and the construction of college English teaching quality assessment system, this paper aims at the application status of data mining in college English teaching quality analysis in recent years, and evaluates the application effect by combining with experimental design.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115932338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1109/ICTech55460.2022.00113
Wenhai Liu, Jian Wu
Traditional 1553B teaching-help software does anything about what should be taught, however, good data not included. We design and implement a 1553B network driven by simulation data flow. And it performs well on 1553B bus.
{"title":"Design of 1553B Network for Teaching","authors":"Wenhai Liu, Jian Wu","doi":"10.1109/ICTech55460.2022.00113","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00113","url":null,"abstract":"Traditional 1553B teaching-help software does anything about what should be taught, however, good data not included. We design and implement a 1553B network driven by simulation data flow. And it performs well on 1553B bus.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121520661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1109/ICTech55460.2022.00060
Li-na Wang, Y. Ren, Fengqin Wang
The realization of intelligent air combat process is a hot research direction of military powers at present. Flight action recognition is also the basis of many key technologies to realize intelligent air combat. Therefore, according to the flight action characteristics of military aircraft, it will be a very valuable work to realize an automatic flight action recognition model based on flight parameter data. In this paper, the CNN model with attention mechanism is used for flight action recognition. The real data generated in the training process are used to experiment the established model, and the output results show that, The model has high accuracy for flight action recognition.
{"title":"CNN Model Based on Attention Mechanism and Its Application in Time Series Data","authors":"Li-na Wang, Y. Ren, Fengqin Wang","doi":"10.1109/ICTech55460.2022.00060","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00060","url":null,"abstract":"The realization of intelligent air combat process is a hot research direction of military powers at present. Flight action recognition is also the basis of many key technologies to realize intelligent air combat. Therefore, according to the flight action characteristics of military aircraft, it will be a very valuable work to realize an automatic flight action recognition model based on flight parameter data. In this paper, the CNN model with attention mechanism is used for flight action recognition. The real data generated in the training process are used to experiment the established model, and the output results show that, The model has high accuracy for flight action recognition.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114940138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-01DOI: 10.1109/ICTech55460.2022.00011
Zhang Yi, J. Ally, Kang Juan, Mei Cong
The intelligent clothing system involves a large number of acquisition processes. In general, this process includes data terminal acquisition, data cleaning, data processing and analysis. These devices can monitor the area covered by the network in real time. Even through the adjustment of system parameters, so as to improve the comfort of clothing. For example, in high temperature environment, through the system control to speed up the loss of body heat. On the contrary, in low temperature environment, through regulating control to reduce the body heat emission, prevent temperature loss. An analysis of body perception parameters is used to predict the outcome of a risk before it occurs. Thus improve the effectiveness of medical rescue. The work is to design an efficient monitoring network for local monitoring and share data with the terminal Numerical simulation proves that the design strategy in this article can improve the network throughput and process the data in the order of priority.
{"title":"Data Classification and Aggregation in Flexible Clothing Based on Cloud Computing Analysis","authors":"Zhang Yi, J. Ally, Kang Juan, Mei Cong","doi":"10.1109/ICTech55460.2022.00011","DOIUrl":"https://doi.org/10.1109/ICTech55460.2022.00011","url":null,"abstract":"The intelligent clothing system involves a large number of acquisition processes. In general, this process includes data terminal acquisition, data cleaning, data processing and analysis. These devices can monitor the area covered by the network in real time. Even through the adjustment of system parameters, so as to improve the comfort of clothing. For example, in high temperature environment, through the system control to speed up the loss of body heat. On the contrary, in low temperature environment, through regulating control to reduce the body heat emission, prevent temperature loss. An analysis of body perception parameters is used to predict the outcome of a risk before it occurs. Thus improve the effectiveness of medical rescue. The work is to design an efficient monitoring network for local monitoring and share data with the terminal Numerical simulation proves that the design strategy in this article can improve the network throughput and process the data in the order of priority.","PeriodicalId":290836,"journal":{"name":"2022 11th International Conference of Information and Communication Technology (ICTech))","volume":"179 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129599025","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：