Benefit from feature presentation with huge parameters and high GPU computing resources, deep convolution neural network has been widely studied in image denoising due to its considerable denoising performance. However, these parameters will consume quantities of memory and computing resources, meanwhile, lots of them are correlated and redundant. We propose a low calculation cost and fast denoising convolution neural network, namely Mirage Net, inspired by the natural phenomenon of mirage. Based on our refraction convolution, which is the combination of depth-wise and point-wise convolution, Mirage Net can reduce parameter redundancy and learn effective presentations from one-layer deeper feature maps by cheap cost linear transformations which will be concatenated with previous feature maps as input of the next convolution layer. We also use alternating training strategy with multi-loss which accelerate the training processing and convergence rate. Our experiments on public datasets show that Mirage Net can achieve higher quality denoised images than DnCNN, and furthermore, the calculation cost is only half of them.
{"title":"Mirage Net: Low Calculation Cost Network for Image Denoising","authors":"Linsong Xu, Pengcheng Ouyang","doi":"10.1145/3421766.3421769","DOIUrl":"https://doi.org/10.1145/3421766.3421769","url":null,"abstract":"Benefit from feature presentation with huge parameters and high GPU computing resources, deep convolution neural network has been widely studied in image denoising due to its considerable denoising performance. However, these parameters will consume quantities of memory and computing resources, meanwhile, lots of them are correlated and redundant. We propose a low calculation cost and fast denoising convolution neural network, namely Mirage Net, inspired by the natural phenomenon of mirage. Based on our refraction convolution, which is the combination of depth-wise and point-wise convolution, Mirage Net can reduce parameter redundancy and learn effective presentations from one-layer deeper feature maps by cheap cost linear transformations which will be concatenated with previous feature maps as input of the next convolution layer. We also use alternating training strategy with multi-loss which accelerate the training processing and convergence rate. Our experiments on public datasets show that Mirage Net can achieve higher quality denoised images than DnCNN, and furthermore, the calculation cost is only half of them.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126573592","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}
It has achieved a consensus that the on-sale vehicles and spare parts should reach certain requirements in the field of safety, energy saving, environment protection and immobilization to guarantee the performance of them. Two systems of ECE and FMVSS in automobile field have formed after development of nearly a hundred years. In this article, we will compare the testing points' differences between lower beam and upper beam in two main mandatory systems in order to promote the lighting distribution design of headlamp.
{"title":"Analyse of Differences between Lower Beam and Upper Beam of ECE and FMVSS","authors":"Tao Zhang","doi":"10.1145/3421766.3421862","DOIUrl":"https://doi.org/10.1145/3421766.3421862","url":null,"abstract":"It has achieved a consensus that the on-sale vehicles and spare parts should reach certain requirements in the field of safety, energy saving, environment protection and immobilization to guarantee the performance of them. Two systems of ECE and FMVSS in automobile field have formed after development of nearly a hundred years. In this article, we will compare the testing points' differences between lower beam and upper beam in two main mandatory systems in order to promote the lighting distribution design of headlamp.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129074859","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}
In recent years, the introduction of intelligent optimization algorithm into cloud computing task scheduling to deal with the problem of massive task scheduling is a research hotspot. This paper proposes three improved differential evolution cloud computing task scheduling algorithms, and the application of the improved differential evolution algorithm in cloud computing task scheduling problem is mainly studied. The maximum task completion time is optimized by improving parameters F, CR, and variation strategies. Through two sets of simulation experiments, it is proved that three improved differential evolutionary cloud task scheduling algorithms have less task completion time than the traditional differential evolution algorithm, and the bigger the number of tasks, the more obvious the performance optimization of the algorithm.
{"title":"Cloud Computing Task Scheduling Based on Improved Differential Evolution Algorithm","authors":"Xueliang Fu, Yumeng Hu, Yang Sun","doi":"10.1145/3421766.3421785","DOIUrl":"https://doi.org/10.1145/3421766.3421785","url":null,"abstract":"In recent years, the introduction of intelligent optimization algorithm into cloud computing task scheduling to deal with the problem of massive task scheduling is a research hotspot. This paper proposes three improved differential evolution cloud computing task scheduling algorithms, and the application of the improved differential evolution algorithm in cloud computing task scheduling problem is mainly studied. The maximum task completion time is optimized by improving parameters F, CR, and variation strategies. Through two sets of simulation experiments, it is proved that three improved differential evolutionary cloud task scheduling algorithms have less task completion time than the traditional differential evolution algorithm, and the bigger the number of tasks, the more obvious the performance optimization of the algorithm.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129102143","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}
As the key to fire attack and defense in offshore war, rapid and accurate positioning through automatic identification of maritime targets has always been the greatest concern of global military research. This paper focuses on an offshore target identification method based on the improved K-means clustering algorithm, which combines the advantages of k-means clustering algorithm of favorable clustering effect, convergence speed and recognition effect. Large-scale offshore target signal source data is converted into digital signals, and the shortest distance between each particle and its corresponding class is obtained by improving the K-means clustering algorithm. The signals are then divided into several different clusters to achieve target identification. According to the results of a practical example, the method demonstrates notable performance and high practical value in the fast and automatic identification of maritime targets.
{"title":"Automatic Identification of Maritime Targets based on K-means Optimization Algorithm","authors":"Guanghui Yin, Jingfei Yang","doi":"10.1145/3421766.3421817","DOIUrl":"https://doi.org/10.1145/3421766.3421817","url":null,"abstract":"As the key to fire attack and defense in offshore war, rapid and accurate positioning through automatic identification of maritime targets has always been the greatest concern of global military research. This paper focuses on an offshore target identification method based on the improved K-means clustering algorithm, which combines the advantages of k-means clustering algorithm of favorable clustering effect, convergence speed and recognition effect. Large-scale offshore target signal source data is converted into digital signals, and the shortest distance between each particle and its corresponding class is obtained by improving the K-means clustering algorithm. The signals are then divided into several different clusters to achieve target identification. According to the results of a practical example, the method demonstrates notable performance and high practical value in the fast and automatic identification of maritime targets.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127728199","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}
Accidental fire phenomenon, that the ignition of one microexplosive chamber element leads to that the adjacent ones are ignited, was a big problem for MEMS(Micro-Electro-Mechanical System)-based solid propellant micro-thruster arrays. In order to solve this problem, thermal protection capabilities of microexplosive chamber were modeled and simulated, and both the rectangle and annular isolation chambers were designed for comparison. The simulation results showed that the Pyrex 7740 glass-based micro-explosive chamber isolated heat better than the Si-based chamber. The maximum temperatures of the chamber adjacent to the central chamber decreased exponentially with the length increasing of rectangle isolation chamber, decreased linearly with the width increasing of rectangle isolation chamber, increased exponentially with the angle increasing of annular isolation chamber, and decreased exponentially with the radius increasing of annular isolation chamber, which were valuable for the design of MEMS-based solid propellant micro-thruster arrays.
{"title":"Thermal Protection Capability of Micro-explosive Chamber for Micro-thruster Array","authors":"Wei Liu, Yan Xue, Ruizhen Xie, X. Ren, Lan Liu","doi":"10.1145/3421766.3421883","DOIUrl":"https://doi.org/10.1145/3421766.3421883","url":null,"abstract":"Accidental fire phenomenon, that the ignition of one microexplosive chamber element leads to that the adjacent ones are ignited, was a big problem for MEMS(Micro-Electro-Mechanical System)-based solid propellant micro-thruster arrays. In order to solve this problem, thermal protection capabilities of microexplosive chamber were modeled and simulated, and both the rectangle and annular isolation chambers were designed for comparison. The simulation results showed that the Pyrex 7740 glass-based micro-explosive chamber isolated heat better than the Si-based chamber. The maximum temperatures of the chamber adjacent to the central chamber decreased exponentially with the length increasing of rectangle isolation chamber, decreased linearly with the width increasing of rectangle isolation chamber, increased exponentially with the angle increasing of annular isolation chamber, and decreased exponentially with the radius increasing of annular isolation chamber, which were valuable for the design of MEMS-based solid propellant micro-thruster arrays.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130047275","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}
The continuous development of photovoltaic grid-connected technology extended the requirement on higher power density and higher efficiency for power converters. In this respect, the application of silicon carbide (SiC) high-power power electronic devices in photovoltaic inverter systems can simplify the system design, simplify the heat dissipation device, reduce energy loss, reduce the volume and weight of the system, and thus greatly reduce the cost. The boost module is a key component of the inverter. In this work, 1200V/20A SiC diodes and SiC MOSFETs are applied to the boost circuit of a single-phase photovoltaic grid-connected inverter, which increases the overall efficiency of the inverter by more than 0.5%.
{"title":"The Application of SiC Devices in Photovoltaic Grid-connected Inverters","authors":"Yongping Li, Ling Li, Hao Wu, Chunyan Jiang, Chaozhang Liu, Shumin Yan, Xinhua Chen, Guodong Feng, Huanyun Zhang","doi":"10.1145/3421766.3421819","DOIUrl":"https://doi.org/10.1145/3421766.3421819","url":null,"abstract":"The continuous development of photovoltaic grid-connected technology extended the requirement on higher power density and higher efficiency for power converters. In this respect, the application of silicon carbide (SiC) high-power power electronic devices in photovoltaic inverter systems can simplify the system design, simplify the heat dissipation device, reduce energy loss, reduce the volume and weight of the system, and thus greatly reduce the cost. The boost module is a key component of the inverter. In this work, 1200V/20A SiC diodes and SiC MOSFETs are applied to the boost circuit of a single-phase photovoltaic grid-connected inverter, which increases the overall efficiency of the inverter by more than 0.5%.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"52 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122422010","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}
This paper analyzes the development and evolution of phoneme-level features, especially language recognition methods based on underlying acoustic features, and Dialect identification technology at home and abroad, which Explains that the research of Dialect identification technology based on language recognition methods consists of feature extraction and model establishment, and highlights the current development of Dialect Identification technology based on i-vector features and neural network modeling methods and continuous breakthroughs in research key and difficult points.
{"title":"Research on the Development of Automatic Dialect Identification Technology Based on Language Recognition","authors":"Yuanyuan Yang, Kailei Li, Rui Guo, Yun Yan","doi":"10.1145/3421766.3421793","DOIUrl":"https://doi.org/10.1145/3421766.3421793","url":null,"abstract":"This paper analyzes the development and evolution of phoneme-level features, especially language recognition methods based on underlying acoustic features, and Dialect identification technology at home and abroad, which Explains that the research of Dialect identification technology based on language recognition methods consists of feature extraction and model establishment, and highlights the current development of Dialect Identification technology based on i-vector features and neural network modeling methods and continuous breakthroughs in research key and difficult points.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132906440","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}
This paper designed a smart car control system based on the digital camera OV7620, and STM32 as a control core using digital cameras to obtain road information. To save CPU resources, it enabled DMA function of the host controller. Further, this chapter describes the method of image filtering, image denoising and center line extraction. DC Brushless motor power motor with speed and current double closed-loop control was used. System equipped with an ultrasonic obstacle avoidance sensor for detection of road barriers. Practical tests show that the intelligent control system has a good autonomy, and can be a good simulation of actual driving pattern.t History (specific to journal submission), and finally, Manuscript validation and placing the respective metadata (Bibstrip/copyright text) while applying the required template.
{"title":"Design of Smart Car Control System Based on Digital Camera","authors":"Rui Xue, Shixian Zeng, Sensen Song","doi":"10.1145/3421766.3421806","DOIUrl":"https://doi.org/10.1145/3421766.3421806","url":null,"abstract":"This paper designed a smart car control system based on the digital camera OV7620, and STM32 as a control core using digital cameras to obtain road information. To save CPU resources, it enabled DMA function of the host controller. Further, this chapter describes the method of image filtering, image denoising and center line extraction. DC Brushless motor power motor with speed and current double closed-loop control was used. System equipped with an ultrasonic obstacle avoidance sensor for detection of road barriers. Practical tests show that the intelligent control system has a good autonomy, and can be a good simulation of actual driving pattern.t History (specific to journal submission), and finally, Manuscript validation and placing the respective metadata (Bibstrip/copyright text) while applying the required template.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130917478","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}
In order to monitor the power purchase and consumption of users in power plants or power companies in real time, this study analyzes the power linkage marketing situation. The research combines the information entropy decision tree model and the clustering algorithm in data mining to realize the visualization of power high-dimensional data, so as to realize the intelligent monitoring of power. The results show that the electricity transaction volume between users and the company accounts for about 40% of the total transaction volume. Data visualization based on information entropy decision trees and clustering algorithms can effectively monitor the purchase and use of electricity by users. It is hoped that this research will provide certain reference and reference for the power marketing of power companies and the monitoring of users' power consumption.
{"title":"Multi-Dimensional Power Marketing Linkage Analysis and Intelligent Monitoring Based on Visualization Technology","authors":"Shaodong Zhao","doi":"10.1145/3421766.3421816","DOIUrl":"https://doi.org/10.1145/3421766.3421816","url":null,"abstract":"In order to monitor the power purchase and consumption of users in power plants or power companies in real time, this study analyzes the power linkage marketing situation. The research combines the information entropy decision tree model and the clustering algorithm in data mining to realize the visualization of power high-dimensional data, so as to realize the intelligent monitoring of power. The results show that the electricity transaction volume between users and the company accounts for about 40% of the total transaction volume. Data visualization based on information entropy decision trees and clustering algorithms can effectively monitor the purchase and use of electricity by users. It is hoped that this research will provide certain reference and reference for the power marketing of power companies and the monitoring of users' power consumption.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116474141","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}
In this paper, the mechanical performance of welded steel joints under different heat treatment methods is studied. Through experiments and finite element analysis, the vulnerable parts are found out. In this paper, the mechanical properties of welded joints of 5A06 aluminum alloy under different processing conditions, including tensile strength and fatigue strength, were studied, and the P-S-N curve of welded joints was drawn based on the test data.In addition, the effect of repair welding on the mechanical properties of the welded joint was studied by tensile test, fatigue test and fatigue fracture morphology analysis.
{"title":"Local Mechanical Property Simulation and Analysis of Aluminum Alloy Welded Joint with Different Welding Processes","authors":"Yicheng Jiang","doi":"10.1145/3421766.3421872","DOIUrl":"https://doi.org/10.1145/3421766.3421872","url":null,"abstract":"In this paper, the mechanical performance of welded steel joints under different heat treatment methods is studied. Through experiments and finite element analysis, the vulnerable parts are found out. In this paper, the mechanical properties of welded joints of 5A06 aluminum alloy under different processing conditions, including tensile strength and fatigue strength, were studied, and the P-S-N curve of welded joints was drawn based on the test data.In addition, the effect of repair welding on the mechanical properties of the welded joint was studied by tensile test, fatigue test and fatigue fracture morphology analysis.","PeriodicalId":360184,"journal":{"name":"Proceedings of the 2nd International Conference on Artificial Intelligence and Advanced Manufacture","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124836021","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：