2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)最新文献

{"title":"Gesture-based Intention Prediction for Automatic Door Opening using Low-Resolution Thermal Sensors: A U-Net-based Deep Learning Approach","authors":"Sheng-Ya Chiu, Sheng-Yang Chiu, Yu-Ju Tu, Chi-I Hsu","doi":"10.1109/ECICE52819.2021.9645718","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645718","url":null,"abstract":"Personal health consciousness has increased amid pandemics. The implementation of automatic doors could help stop the infection. The need for an intelligent sensor emerges for automatic doors to prevent unneeded open as well as customer privacy concerns. This research proposes a novel automatic door opening mechanism using a low-resolution thermal sensor, based on which a multi-task U-Net structure network is adopted to classify hand-raising gestures. With the aid of segmentation masking, there is 74% reduction of training steps for convergence than that of mere thermal image classification while maintaining similar classification performance. On-site deployment of this approach via constantly collecting door-opening misclassification cases for model improvement will lead to practical success in the near future.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125817810","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":"[ECICE 2021 Front matter]","authors":"","doi":"10.1109/ecice52819.2021.9645685","DOIUrl":"https://doi.org/10.1109/ecice52819.2021.9645685","url":null,"abstract":"","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114251778","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 on Humidity Status Fuzzy Estimation of Low-power PEMFC Stack Based on the Softsensing Technology","authors":"Litian Zhang, Caijun Rao, Chengxu Huang, Beitian Zheng, Sheng-Feng Lin, Wenyu Zhang, Jiaqi He, Baohua Tan","doi":"10.1109/ECICE52819.2021.9645631","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645631","url":null,"abstract":"Internal humidity is an important parameter and strongly influences the life and performance of proton exchange membrane fuel cells. However, due to the particularity of the closed structure of the fuel cell, the existing tools and methods cannot directly measure it. Focusing on the low-power fuel cells stack, a method for estimating the humidity of fuel cells based on soft-sensing technology is proposed in this paper. After being combined with the basic concept of fuzzy mathematics, three parameters are taken as the input value of the fuzzy logic soft-sensing model, including the internal resistance, the sum of the initial open-circuit voltage and the battery voltage under the given load current, and the difference between the initial open-circuit voltage and the battery voltage under the given load current. Then the soft-sensing technology model has been established and trained, and the actual runtime data has been adopted to estimate the humidity status. The experimental results proofed and verified the method based on the soft-sensing technology.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"150 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114475687","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":"Mining Money Transaction Path Based on Graph Computing with fuzzy Association Constraints","authors":"Jianying Xiong, H. Gong","doi":"10.1109/ECICE52819.2021.9645617","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645617","url":null,"abstract":"With the rapid increase of economic crimes and the renovation of technology, it is a challenge to analyse the behaviour of capital transactions. A key problem is to infer the path of the capital transaction from the transaction network. Through the network computing theory, we evaluate the risk of the trading node by taking the node as the node weight of the path according to the risk value. According to the directivity of the path of fund transaction, we construct the constraints of account transaction association conditions, calculate the weight of fund transfer path under different constraints, and propose a network graph model combining the characteristics of node-set and transaction constraint rules. Compared with the traditional manual and fixed threshold mode, the model realizes the reasonable reasoning of the transaction path and provides technical support for illegal fund tracking.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123955047","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":"[Copyright notice]","authors":"","doi":"10.1109/ecice52819.2021.9645715","DOIUrl":"https://doi.org/10.1109/ecice52819.2021.9645715","url":null,"abstract":"","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121466056","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":"RPA and L-System Based Synthetic Data Generator for Cost-efficient Deep Learning Model Training","authors":"E. S., O. E. Ramos, Sixto Prado G.","doi":"10.1109/ECICE52819.2021.9645719","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645719","url":null,"abstract":"Deep learning (DL) models applied to computer vision have made great progress for image-based plant phenotyping in recent years, mostly for quality control process automation in the agroindustry. On the one hand, these models are able to detect objects in complex and noisy images as fast as human observations, but on the other hand, they are trained with a large amount of labeled data for parameter tuning. This turns the training process into an expensive, repetitive, and time-consuming labor. In this work, a synthetic data generator based on robotic process automation (RPA) and Lindenmayer systems (L-Systems) named RPASD is designed and implemented to train a DL model that detects artichoke seedlings in images captured by a robot. First, the growth artichoke seedling is modeled in L+C language using the LStudio software. Second, the RPASD is developed in Python to produce labeled images of grouped synthetic artichoke seedlings that alongside manually labeled images of real artichoke seedlings, taken by a robot, form the PlantiNet database. Third, a YOLOv3 model is trained with the previously built databases forming three datasets: 1) real and synthetics seedlings, 2) only synthetic seedlings, and 3) only real seedlings. The results show a 55% of Mean Intersection over the Union (mIoU) when training only with the second dataset and testing with the third one, which allows us to conclude that our proposed method could adequately boost DL model training reducing costs and time.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121574481","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":"Parametric Analysis of a High-efficient Heat Exchanger","authors":"Kai-Cheng Hsu, Chiu-Feng Lin, C. Tsai, T. Chen, Zi-Cheng Liu, Y. Hong","doi":"10.1109/ECICE52819.2021.9645603","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645603","url":null,"abstract":"Due to the increasingly serious impact of environmental pollution in recent years, people’s awareness of environmental protection has also increased. Owing to the shortage of petrochemical fuels and the increase in prices, people are looking for alternative energy sources, while also trying to use renewable energy to reduce energy consumption rate and greenhouse gas emissions. According to the International Energy Agency (IEA) research on renewable energy and energy conservation technologies, it is expected to reduce 1.5 billion tons of CO2 Emissions by 2050 if people utilize energy regeneration technology and continue to improve its regeneration efficiency. Nowadays, many industries often use heat exchangers to recover heat energy, such as the petrochemical industry, steel industry, metal processing industry, and so on. When it comes to designing the heat exchanger, we need to consider its material and size and evaluate its heat transfer efficiency, pressure loss, and production cost. In this study, three types of heat exchangers with different geometrical shapes and fins are designed, which are flat fins, tail fins, and eye-shaped fins. The ANSYS/ FLUENT software is used to build models and simulate, mainly for the above three types. The heat transfer efficiency and pressure loss of heat exchangers with different geometrical fins are discussed. After simulation and comparison, the heat transfer efficiency of the flat fin is the best among the three types of fins, but the pressure loss is the largest, while the heat transfer efficiency of the eye-shaped fin is slightly lower than that of the flat fin, and the pressure loss is the smallest among the three fins. This study divides the enthalpy value of the air on the cold side by the pressure loss of the exhaust gas on the hot side and uses it as an index for comparing heat exchangers’ performance. The design of the heat exchangers is better if the index is higher. Among the three types of fins, the eye-shaped fins’ index value is the highest, followed by the tail-type fin, and the flat-type fin has the lowest value.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126011622","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":"Combination of BP Neural Network and Logistic Regression its Application","authors":"Lei Wei, Yao He","doi":"10.1109/ECICE52819.2021.9645605","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645605","url":null,"abstract":"Both BP neural network and logistic regression are widely applied in the field of nonlinear relationship analysis. This paper combines the logistic regression model and BP neural network for small sample prediction to establish a new nonlinear fitting model and apply it to practice. The new model effectively extracts the main control variables under multi-factor interference. The accuracy of the prediction model is further improved, which is highly consistent with the significance test of logistic regression.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"181 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132224586","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":"Development of the High Voltage DC Power Supply for X-ray Tube","authors":"Yongfan Pu, Xubo Wei, Zhao Liu","doi":"10.1109/ECICE52819.2021.9645670","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645670","url":null,"abstract":"A 350 kV/20 mA high voltage DC power supply for X-ray tube is designed to use as a control circuit with Digital Signal Processor (DSP). The power use the 380 V/50 Hz three-phase electric as input signal, and output high voltage via rectifier filter, full bridge inverter, transformer booster and voltage doubling rectifier. The inverter circuit implements voltage adjustment by generating SPWM of different duty ratios in DSP. The Maxwell software is employed to calculate the distribution of the electric field as designing the transformer, and the maximum field strength is 78.521 kV/mm in winding. It is available to design the layer insulation of transformer windowing with the polyimide film in accordance with the electric field. Voltage doubling rectifier circuit uses C - W positive and negative two-way voltage doubling rectifier circuit which is a wide application in X-ray Optical Source. The numerical simulation indicates that the overall design of power is legitimate and the parameters can reach the expected requirements.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134095630","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":"Different Concentrations of Carbon Nanotubes/Graphene and TiO2 Composite Photoanodes for Dye-sensitized Solar Cells","authors":"Liangbo Peng, T. Wu","doi":"10.1109/ECICE52819.2021.9645717","DOIUrl":"https://doi.org/10.1109/ECICE52819.2021.9645717","url":null,"abstract":"Dye-sensitized solar cells have the advantages of low material cost, easy process, and simple process equipment as one of the future green energy developments. As graphene has good electrical conductivity, thermal conductivity, high light transmittance, and low resistance, it is chosen as a photoanode material. Carbon nanotubes have high conductivity, high chemical stability, and excellent mechanical strength, and are suitable for dye-sensitized solar cells. In this study, by adding different concentrations of single-layer graphene and multi-layer carbon nanotubes to titanium dioxide, the effects of graphene and carbon nanotubes on the dye-sensitive battery under different concentrations were investigated. Doping graphene and carbon nanotube dye-sensitive cells in titanium dioxide is measured and compared with the dye-sensitive cells of graphene, carbon nanotubes, and carbon nanotubes/ graphene to analyze the three different processes. Dye-sensitive battery characteristics. The results of the study show that the carbon nanotube/graphene dye-sensitive battery at the same time, compared with the thin film coating doped only with carbon nanotubes and graphene. The surface of the working electrode is rough, which increases the light absorption rate. The increase in surface pores has increased the dye absorption, and the overall light conversion efficiency of the battery has been significantly improved.","PeriodicalId":176225,"journal":{"name":"2021 IEEE 3rd Eurasia Conference on IOT, Communication and Engineering (ECICE)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115547329","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}