{"title":"Human Pose Detection Based on Multi-scale and Multi-stage Structure Network","authors":"Yalan Li, Jiangquan Huan, Xiaoqin Zhang, Min Yao","doi":"10.1109/ICAA53760.2021.00126","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00126","url":null,"abstract":"Human pose detection is essential for human behaviour recognition. In this paper, an end-to-end network is proposed to improve the precision of pose detection by fusing multi-scale features, detected parts of the human body at coarse stages and the detected human positions. The network parameters are trained uniformly by Adam optimization algorithm. A large quantity of experiments is carried out on MPII dataset and the images captured by ourselves to explore the relationship between the detection rate and the network structure. The experimental results show that the network can achieve a high detection rate.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114177465","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}
{"title":"Overview of the Basic Principles of Blockchain","authors":"Buquan Liu","doi":"10.1109/ICAA53760.2021.00108","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00108","url":null,"abstract":"In the Internet, the authenticity of data between computer nodes is difficult to be guaranteed. However, as the underlying implementation of bitcoin, blockchain adopts decentralized and distributed ledger to allow false transactions and data tampering, and ultimately provides a reliable and trustworthy solution for bitcoin transactions on the unreliable and untrusted Internet. This paper introduces the basic principles of blockchain from the aspects of the relationship between blockchain and bitcoin, two mathematical algorithms involved in blockchain, the logical structure of blockchain, the types of blockchain, and the application of blockchain. This paper explains the core concepts of blockchain, such as decentralization, trustlessness, mining, proof of work, points out several problems that need to be paid attention to in the practical development of blockchain application, and puts forward the approach of integrating blockchain and traditional database to solve the practical application.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"4 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114264567","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}
{"title":"Ship Target Classification Technology Fused With Domain Knowledge","authors":"Z. Luo, Zhou Fang, Shuai Ding","doi":"10.1109/ICAA53760.2021.00092","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00092","url":null,"abstract":"For traditional convolution neural network using general classify convolution kernels, due to low resolution target is difficult to effectively identify problems, this paper proposes a ship target classification method of domain knowledge, using the port latitude and longitude of the location, its often parked ship types are divided into groups, makes the model in the corresponding category judgment within the group to join ship size and other information at the same time, by comparing the information, to correct the ship classification, to enhance the model fine classification ability. By adding prior knowledge, the accuracy of ship identification is improved.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"417 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114429076","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}
{"title":"Construction of Mobile Service Platform in University Library","authors":"Feng Li, K. Liao","doi":"10.1109/ICAA53760.2021.00065","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00065","url":null,"abstract":"In recent years, with the popularity of wireless networks, mobile library as a quick and convenient service is used in the construction of university library, but its service mode is single and hard to better play the advantages of mobile service. In this paper, a kind of based on the integration of functions, basic services, extended services and personalized service composition of mobile service platform in university libraries. At last, the case of mobile service platform construction of Huazhong Agricultural University Library is introduced.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126832200","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}
{"title":"Research on Command and Control of Maritime Emergency Rescue Based on Big Data","authors":"Lihong Qu, Ruolin Xing, Jiajun Hou","doi":"10.1109/ICAA53760.2021.00105","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00105","url":null,"abstract":"Maritime big data emergency rescue command and control is based on the concentration and integration of public security and national marine big data. Through mining, forecasting and evaluating relevant big data, it can provide security early warning, prepare emergency rescue plan and implement emergency rescue plan in real time. Based on the application of big data, this paper discusses the capability requirements and basic concepts of maritime emergency rescue command and control based on big data, and analyzes the information flow and key technologies of big data emergency rescue command and control.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124385843","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}
Qiang Han, Xianguo Tuo, Yongshuai Zhou, Yuxin Tang
{"title":"Consensus Problem of Leader-following Second-order Multi-agent Systems with Time-delay under Directed Topology","authors":"Qiang Han, Xianguo Tuo, Yongshuai Zhou, Yuxin Tang","doi":"10.1109/ICAA53760.2021.00036","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00036","url":null,"abstract":"In this paper, consensus problems of a kind of continuous-time second-order MAS with communication delays is studied. A neighbor-based linear consensus protocol with time delay is introduced based on the fixed communication topology firstly, and then a necessary and sufficient condition for the system is derived based on graph theory and quadratic stability theory. An equality is proposed for obtaining the maximum time delay which is induced in the positions and velocities. Numerical simulations are carried out to further illustrate the effectiveness of the proposed theoretical results.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121519639","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}
{"title":"Design and Implementation of a Chatbot System Integrated with Facial Expression Recognition","authors":"Sijing Hu, Chengzhang Qu","doi":"10.1109/ICAA53760.2021.00087","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00087","url":null,"abstract":"Different from those who desires to talk, those “empty-nest youth” who do not take the initiative to talk about their emotions still want someone to be with them, and interacting with them, give them some care, although they cannot talk about their emotions. In order to pay attention to the mental health status of this group of people and meet their inner demands, this paper first uses facial expression recognition technology to analyze users' emotions in real time. Then employs chatbot technology to intervene the user's mood through automated responses. In the process of realizing the chatbot function, MongoDB is introduced to store the sayings of chatbots with different emotions, and seven chatbots with different emotions are divided. In addition, MongoDB is used to store the scores of users' emotions, which can achieve the function of recording users' emotions. Finally, Django framework is used to realize the page of the web side, which can call all the functions of the system and achieve the effect of different interventions according to the user's emotions.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121793375","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}
{"title":"Point2Token: A Multi-Tagging Answer Retrieval Framework for Question Answering","authors":"Yi Liu, Puning Yu","doi":"10.1109/ICAA53760.2021.00069","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00069","url":null,"abstract":"Question answering plays a crucial role in the chatbot systems, in which it retrieves the answer from the given context and return the predicted span as a result to users. Previous work mostly modelled this task as a multi-classification problem. However, the models cannot gain a promising result due to the scarcity of the probability distribution over the whole given context. In this paper, we propose a novel approach to solve the problem mentioned above. We model the question answering task as a multiple binary classification problem and introduce PointerNet in our model decoder to predict whether it belongs to a start or end position in each token within context. The experimental results on a well-studied dataset show that our model outperforms the baseline models, which proves our model effectiveness.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131700259","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}
Xingda Sun, Caixia Lu, Jia He, Yuan Liu, Yuanyuan Zhao, Yimeng Wang, Ye Du
{"title":"Realization of Visualization of Character Relationship Based on Relationship Graph","authors":"Xingda Sun, Caixia Lu, Jia He, Yuan Liu, Yuanyuan Zhao, Yimeng Wang, Ye Du","doi":"10.1109/ICAA53760.2021.00147","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00147","url":null,"abstract":"Through big data analysis, a multi-level and multi-dimensional tag database of leading cadres has been constructed. Based on the statistics of the characteristics of data information, establish staff position tags, organization tags, and key event tags for common information to achieve rapid screening and analysis of data information. Finally, the four relationship dimensions of the personnel are displayed in the form of data visualization, namely, former colleagues, current colleagues, fellow villagers and alumni.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131857865","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}
{"title":"Text Similarity Based on Post-training BERT","authors":"Yongping Xing, Chaoyi Bian","doi":"10.1109/ICAA53760.2021.00058","DOIUrl":"https://doi.org/10.1109/ICAA53760.2021.00058","url":null,"abstract":"Text similarity is an important taskin natural language processing. The pre-training BERT model which is get from through large-scale corpus traininghas achieved good results in various natural language processing tasks. However, domain knowledge is not introduced to the model. After the post-training through domain data, the bias of the model for the domain knowledge will be reduced, which improves performance in reading comprehension and emotional aspect extraction. In this paper, the domain knowledge is introduced through the post-training andthen text similarity is discussed.","PeriodicalId":121879,"journal":{"name":"2021 International Conference on Intelligent Computing, Automation and Applications (ICAA)","volume":"241 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120938681","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}