Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers最新文献

{"title":"Improved model for knowledge representation: TransMR","authors":"Xuyang Wang, Yiyuan Zhang","doi":"10.1145/3544109.3544394","DOIUrl":"https://doi.org/10.1145/3544109.3544394","url":null,"abstract":"Knowledge graphs have developed rapidly in recent years, and knowledge representation learning is a fundamental task of knowledge graphs, so knowledge representation learning has likewise received widespread attention. Therefore, a series of knowledge representation models based on TransE methods have been proposed by scholars one after another, among which, the translation model TransE has low model complexity, high computational efficiency, and strong semantic representation ability for knowledge representation of triples. However, the TransE method cannot handle dealing with complex relations. In view of this, this paper proposes an improved knowledge representation model TransMR based on the TransE method, which uses the Marxian distance instead of Euclidean distance to calculate the distance between vectors, and builds entity and relationship models in entity space and relationship space respectively, in which the back propagation and nonlinear operation of single layer neural network are used to enhance the semantic connection between them. Meanwhile, during the model training process, experiments are conducted to improve the fault tolerance of negative example triples by using substitution for the most similar entities.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127851853","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":"The RBFNN Adaptive Variable Structure Temperature Control Algorithm for Extruder with Saturated Input","authors":"Bo Xu, Xiumei Chen, Yufei Qin","doi":"10.1145/3544109.3544186","DOIUrl":"https://doi.org/10.1145/3544109.3544186","url":null,"abstract":"As an important industrial equipment, extruder has high requirements for temperature control accuracy, interference between temperature zones, limited control input, difficult parameter adjustment and complex controller design. Taking extruder temperature control system as the research object, this paper designs extruder temperature control algorithm under the condition of limited input. The algorithm adopts adaptive neural network to automatically identify the system model and suppress the disturbance through convenient structure control algorithm, at the same time, the neural network is used to compensate the saturated input signal. The simulation results show that the algorithm is reliable.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121300193","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":"Multi-feature Microblog Sentiment Analysis based on BERT-AttBiGRU model","authors":"Xuyang Wang, Na He","doi":"10.1145/3544109.3544391","DOIUrl":"https://doi.org/10.1145/3544109.3544391","url":null,"abstract":"In recent years, there has been an increasing amount of research on Weibo sentiment analysis techniques, but less attention has been paid to emoji. However, emoji are closely related to Weibo sentiment. So to judge the microblog sentiment tendency more accurately, in this paper, we select Weibo comments that contain a large number of emoji and propose a neural network classification model that combines emoji. The model first obtains word vectors containing contextual semantic information through the BERT pre-training model, then extracts deep-level feature information by using bi-directional gated recurrent network (BiGRU), and then puts the emoji vector and text vector into the Attention mechanism, and assigns weights to the extracted feature information to highlight the important information. Finally, the Softmax function is used to classify the microblog sentiment.The experimental results prove that the accuracy of the model reaches 97.65%, which effectively improves the accuracy of microblog sentiment classification.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129013690","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 the application of enterprise portal integration technology","authors":"Yali Tian, Qinghe Wo, Lei Zhang","doi":"10.1145/3544109.3544309","DOIUrl":"https://doi.org/10.1145/3544109.3544309","url":null,"abstract":"In view of the business needs and construction status of system applycation and integration of manufacturing enterprises, this paper proposes a solution for digital workshop data bus application based on message middleware. Through the research and implementation application of data bus technology in enterprise portal integration, product design, manufacturing, production and other related application systems are integrated with business, data and applications, and a collaborative management platform with information and data sharing and smooth business processes is built for manufacturing enterprises.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116972964","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":"Student Pilot Performance Evaluation Based on MASPP Model","authors":"Liang Ye, Yu Qian","doi":"10.1145/3544109.3544121","DOIUrl":"https://doi.org/10.1145/3544109.3544121","url":null,"abstract":"At present, the evaluation of student pilots mainly relies on the way of teacher rating, which has the problem of incomplete or redundant indicators, which is not conducive to the comprehensive evaluation. To explore the student pilots in the whole life cycle of the pilot skills in the preliminary phase of one of the management system of performance evaluation model, based on the model of MAPP student pilots license in the airline business is studied according to the factors affecting the performance of this phase in the process of composition, reference the pilot core competence indicators, developed for student pilots MASPP initial evaluation system, Using the method of questionnaire investigation and factor screening large principal component influencing factors, including knowledge, attitude, communication, applications, automation of five dimension principal component and two dimensions including teamwork, work load management of principal components, including psychological competence principal component and situational awareness of main components, the questionnaire has good validity.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117113296","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":"Study of flow control strategies combined between suction slot and vortex generator on an axial compressor cascade","authors":"Fengming Li, Shan Ma, Xiaolin Sun","doi":"10.1145/3544109.3544122","DOIUrl":"https://doi.org/10.1145/3544109.3544122","url":null,"abstract":"To improve the compressor performances, in this research, we used numerical simulation software to analyse the effects of a combined strategies between the suction slot and vortex generator, and the influence of the aspiration flow path configuration was further investigated. The calculated results show that the 1% suction mass-flow ratio is obtained by the backward extension configuration with lower suction power, as well as total pressure loss shows a decreasing trend as the increasing amplification ratio at design condition. However, the superiority disappears at the +7 º incidence condition, the total pressure loss and suction power shows increasing trend. Moreover, the aspiration flow paths with a both sides extension configuration shows a significant advantage in suction power economy that is decreased by 0.9% and 1.5% at the -1 º and +7 º incidence condition respective.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114709772","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 and Implementation of Intelligent Financial Audit System Based on Data Mining and Cloud Computing","authors":"Junhong Gao, Yuanxing Zhao, Xiu Wang","doi":"10.1145/3544109.3544340","DOIUrl":"https://doi.org/10.1145/3544109.3544340","url":null,"abstract":"The application and integration of mega data and cloud accounting technology not only changes the mode of enterprise accounting informatization construction, but also improves and promotes the traditional audit work. In the mega data environment, enterprise business processes are gradually increasing, and structured, unstructured and semi-structured data are becoming increasingly complex. It is difficult for a single financial data to effectively reflect the operational status of an enterprise, and it is urgent for auditing to comprehensively and thoroughly analyze the actual situation of an enterprise by comprehensively utilizing multiple data. Under the background of cloud computing as the foundation and mega data as the business engine, the content, mode and risk of traditional audit have changed, and an innovative model of \"cloud audit\" for mega data has emerged. This paper analyzes the impact of mega data and cloud accounting technology on audit from audit scope, audit data, audit risk, audit technology and auditors, establishes an audit implementation framework based on cloud accounting in mega data environment, and expounds the audit business development process under this framework.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126078920","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":"Unsteady Flow Field Investigations at Near Stall for a Transonic Compressor Rotor by Proper Orthogonal Decomposition","authors":"Z. Liu","doi":"10.1145/3544109.3544112","DOIUrl":"https://doi.org/10.1145/3544109.3544112","url":null,"abstract":"The unsteady numerical simulations are conducted to research the tip clearance flow characteristic at near stall of a transonic compressor rotor. Based on the simulation results, decompose the flow fields by proper orthogonal decomposition (POD). Itsuccessfully captures theoscillation frequency 6.83kHz of the tip flow field with huge scale oscillation structures at blade leading and pressure side reveal the unsteady flow field nature. Through the reconstructed flow fields, it finds that the oscillation behaves as a circumference wave in the blade passages.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"13 6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126095948","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 Automatic Tracking Method of Marker Points in Sports Image Sequence Based on Feature Matching","authors":"Wenlong Cheng","doi":"10.1145/3544109.3544366","DOIUrl":"https://doi.org/10.1145/3544109.3544366","url":null,"abstract":"Generally speaking, the methods of automatic tracking and recognition in sports image sequence analysis can be divided into two categories: first, template matching method, which compares each template image with all sub-images in the search area, finds out the most similar sub-image and makes the sub-image a new template, and repeats the above process in the corresponding search area of the next adjacent image; second, feature matching method, which compares the features of the sub-images in the search area and finds the seal. Sports image collection is dynamic, imaging is vague, and the distribution structure of dynamic feature marks is complex. Automatic tracking of dynamic feature marks in sports scenes is the key to realize moving image recognition. In this paper, feature matching is introduced into the automatic tracking method of mark points in sports image sequence. The basic condition that the image to be registered is collinear with the corresponding line segment on the reference image is used to establish the image deformation model, and the full automation of sequence image registration process is realized through the automatic extraction and automatic matching of line segment features.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124899123","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":"Intelligent Route Planning Algorithm based on Genetic Neural Network","authors":"Yi-Zi Ning, Chongjun Yang","doi":"10.1145/3544109.3544405","DOIUrl":"https://doi.org/10.1145/3544109.3544405","url":null,"abstract":"With the development of modern industry towards large-scale and integration, the production process is becoming more and more complex. The process is seriously nonlinear, time-varying, uncertain and the strong combination of variables, which makes many systems lack accurate mathematical description and difficult to analyze and control with traditional theoretical methods. Therefore, it is necessary to study new intelligent control strategies. Real-time and efficient solution of the optimal path in a large-scale road network is a research difficulty in the field of dynamic path induction. During path planning, the robot's own sensors are required to continuously collect and analyze environmental data, so that the robot can find the target point and update it continuously path. Aiming at the shortcomings of the basic GA, such as low efficiency, when calculating the optimization problems of large-scale networks. In this paper, an intelligent route planning algorithm based on genetic neural network is proposed. The environmental information is obtained by five sensors loaded on the front end. The obtained obstacle, pose and target information are used as the input of neural network, and then the weights of neural network are trained and adjusted by GA. Finally, the output of neural network after training and adjustment is used as the driving control force of robot. On this basis, referring to some conclusions of fixture verification, the stability and deformation characteristics of the workpiece are simulated through some parameters, and the GA is used for combinatorial optimization to determine the optimal positioning point. The algorithm proposed in this paper has the advantages of simple calculation and fast convergence, can avoid some local extremum, and the planned collision free path reaches the shortest collision free path. Finally, through the experimental simulation of the algorithm, the results show that the proposed intelligent route planning algorithm based on genetic neural network is correct and effective. In addition, the real-time performance and rapidity are better than the basic GA, and the balance problem of solving efficiency and solving quality in large-scale road network is also solved.","PeriodicalId":187064,"journal":{"name":"Proceedings of the 3rd Asia-Pacific Conference on Image Processing, Electronics and Computers","volume":"444 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124273180","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}