{"title":"Research on Mural Restoration Method Based on Generative Multi-column Transformer","authors":"Hui Ren, Fanhua Zhao, Z. Li, Ke Sun, Shilin Gao","doi":"10.1109/IMCEC55388.2022.10020135","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10020135","url":null,"abstract":"Ancient mural are splendid artistic treasures of ancient civilizations, but their blurring and destruction are becoming more and more serious due to the effects of time and environment. In this paper, a Generative Multi-column Transformer method is proposed to recover the missing areas of mural. The method synthesizes different image components in a parallel manner by transformer and enhances local details by employing implicit diversified Markov random fields regularization. Generative Multi-column Transformer is used in combination with reconstruction and Markov random field loss methods to propagate local and global information from the background to the target defaced region to reconstruct the loss in that part. The experimental results show that the method based on Generative Multi-column Transformer adopted in this paper is able to restore mural images with complex textures and missing areas very well, while the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) of the method are improved, this paper provides theoretical guidance for the digital restoration of murals.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116460182","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":"Point Cloud Segmentation Algorithm Based on Deep Learning and 3D Reconstruction","authors":"Chunchao Dong, Naijian Chen, Yahui Li, Jiawei Bao","doi":"10.1109/IMCEC55388.2022.10019815","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019815","url":null,"abstract":"In order to accurately reflect the surface defects of parts in three-dimensional space, this paper proposes a defect point cloud segmentation algorithm based on deep learning. The system reconstructs the detected object based on binocular vision, and establishes the point image mapping relationship between the color image and the 3D object point. In addition, according to the point image mapping relationship, the image segmentation and point cloud segmentation algorithm are used to segment the defect area, and the corresponding three-dimensional point cloud of the defect area is obtained. This paper uses YOLACT 's deep learning network to intelligently identify defect types and locations to ensure the detection accuracy and real-time performance of the system. In the process of defect detection, the detection image which only contains the defect area is generated according to the segmentation results of the defect image, which not only improves the detection accuracy, but also improves the detection speed.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122470448","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}
Qiantu Zhang, Yaohua Liu, Huan Wang, Shanliang Ren, Songyang Li
{"title":"Semi-supervised Linear Local Tangent Space Alignment and Application to Pattern Recognition","authors":"Qiantu Zhang, Yaohua Liu, Huan Wang, Shanliang Ren, Songyang Li","doi":"10.1109/IMCEC55388.2022.10019962","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019962","url":null,"abstract":"Aiming at the fact that linear local tangent space alignment (LLTSA) is an unsupervised dimension reduction algorithm, semi-supervised linear local tangent space alignment (SS-LLTSA) dimension reduction algorithm is proposed in this paper. In SS-LLTSA, the distance between different points is adjusted by utilizing part class label information, thereby a new distance matrix is formed and the neighborhood is construct through this new distance matrix. The improved method realized the combination of data intrinsic manifold structure and class label information, and more discriminative low-dimension features can been obtained. The effectiveness of the proposed method was verified through several benchmark datasets.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"99 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122625311","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":"Clothing Recognition Based on Improved ResNet18 Model","authors":"Yiwei Dai, Jun Tao, Cong-yu Ouyang, Xudong Wang","doi":"10.1109/IMCEC55388.2022.10020122","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10020122","url":null,"abstract":"At present, there are many researches on image classification and recognition, among which clothing recognition has more important application and research value in life. Image classification generally adopts convolutional neural network training and recognition. Convolutional neural network has better recognition performance, which is better than traditional machine learning algorithm and BP neural network. Convolutional neural network has various network models such as LeNet, AlexNet, VGG and ResNet. This paper takes clothing pictures as objects and proposes a clothing recognition method based on the improved ResNet18 model. The improved ResNet18 through experiments has an accuracy of 94.78% in identifying clothing, which is 1.01 percentage points higher than the original ResNet18 model, and is also higher than the accuracy of VGG and AlexNet, which can be well qualified for clothing recognition tasks.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"236 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122956031","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":"A Multi-task Neural Network Approach for Intelligent Vehicle Environment Perception","authors":"Cen Ming, Zhouyu Cong, Xuefeng Chen, Yang Yan","doi":"10.1109/IMCEC55388.2022.10019908","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019908","url":null,"abstract":"In recent years, driverless cars have become a research boom, and vision perception is an integral part of driverless systems, which requires a number of perceptual tasks such as target detection and lane segmentation. If each task is run separately, it will consume too much hardware resources and ignore the correlation between the tasks. To address the above problems, this paper proposes a multi-task neural network GaYNet for intelligent vehicle environment perception, which is based on the YOLOv4-tiny framework, optimizes the CSPDarknet53-tiny backbone network, and combines the CBAM attention module and the FPN+PAN bidirectional feature pyramid structure to significantly reduce the consumption of computational resources without re ducing the original network detection accuracy. To evaluate the performance of GaYNet model, this paper uses the BDD100K public dataset for training and testing, and the results show that the detection mAP and segmentation mIOU of this network are 75.3% and 35.46%, respectively, and the detection rate on Jetson TX2 can reach about 18 fps after accelerated by TensorRT, which has reached the practical application requirements and improves the perception system for it has reached the practical application requirements, which plays a positive role in improving the operational efficiency of the sensing system.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122017894","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":"Feature selection analysis of systematic review citation screening algorithm","authors":"Shi Li, Yi-Cun Du, Yanhong Yun, Hong-Guang Liu","doi":"10.1109/IMCEC55388.2022.10020072","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10020072","url":null,"abstract":"Recently, increasing studies turned the spotlight on machine learning-based automated citation screening for systematic review. Moreover, outcomes of the model were suggested to be highly influenced by the selection of bibliographic information. In the current study, we implemented automated citation screening algorithm based on literature similarity, and further evaluated the effects of single or combined feature engineering constituted by the title, abstract and publication type. In the single feature-based strategy, we found that abstract-based model showed the highest Work saved over sampling at 95% recall (WSS95) score (73.55%). In terms of multiple features-based approaches, the model with features pooling title, abstract, and publication type showed the highest WSS95 score (77.34%). The current findings suggest that literature information might be mainly contained in abstract and title, thus highlighting their crucial roles in systematic review citation screening.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"34 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116599111","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}
Fengbin Zhang, X. Liu, Zechen Wei, Jiali Zhang, N. Yang, Xuri Song
{"title":"Key Feature Mining Method for Power-Cut Window Based on Grey Relational Analysis","authors":"Fengbin Zhang, X. Liu, Zechen Wei, Jiali Zhang, N. Yang, Xuri Song","doi":"10.1109/IMCEC55388.2022.10019909","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019909","url":null,"abstract":"In the process of compiling the power-cut window period of the power grid equipment maintenance plan, problems such as omission of constraints are prone to occur due to excessive reliance on manual experience. In response to these problems, this paper proposes a method for mining key features of the power-cut window based on grey relational analysis. Through mining and analysis of the historical operation data of the power grid, the operation data of new energy, and the historical power-cut information of equipment, the indicators that play a key role in the arrangement of the outage window period of the equipment maintenance plan are found. Then use the key indicator information to formulate the window period. By mining the relationship between power grid operation data and equipment power outages, this paper can give full play to the big data advantages of the power grid, improve the accuracy and efficiency of the power-cut window period.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129675905","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}
Shuai Li, Chengxin Du, Hui Wu, Yukun Meng, Zhifei Wang
{"title":"Research on Platform Door Control Unit for Intelligent Operating and Maintaining","authors":"Shuai Li, Chengxin Du, Hui Wu, Yukun Meng, Zhifei Wang","doi":"10.1109/IMCEC55388.2022.10020147","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10020147","url":null,"abstract":"In recent years, in the rail transit industry, new technologies represented by fully automatic operation have been emerging one after another, and various new systems and products have been emerging [1]; Users' demand for refined and intelligent operation and maintenance become stronger; If the failure of platform door system during operation cannot be recovered in time, it will have a great impact on the operation. On the other hand, as an important control component of the existing platform door system, door control unit (DCU) has many different options, but also has many problems in operation and maintenance, such as opaque operation information, difficulty in fault location [2], and the maintainability of the system seriously depends on the technical level of the operation and maintenance personnel. To solve the above problems, this paper proposes an intelligent platform door control unit solution, which configures a human-computer interface on the DCU, improves the visual accuracy of operation and maintenance to a single unit of platform doors, achieves the real-time display of equipment status information, fault information, and health information, and provides equipment real-time maintenance strategies in combination with expert maintenance experiences and intelligent analysis to reduce the dependence of equipment operations and maintenances on personal professional level, and improve equipment maintainability","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128527622","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":"Performance Evaluation and Optimization Method in QKD Network for Digital Power Grid","authors":"Zhou Peng, Qi Weiqiang, Xu Min, Xu Zichao, Chen Julong, Wu Hao, Lv Yuxiang","doi":"10.1109/IMCEC55388.2022.10019907","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019907","url":null,"abstract":"In the actual operating environment, the data traffic of each quantum VPN tunnel is different, which will affect the amount of key consumed by each service data encryption. In this work, the key generation rate, bandwidth bottleneck and delay generation mechanism of the transmission link of the multi-relay network node of service packets is analysed. We propose an end-to-end available key rate measurement method by analyzing the end-to-end delay for trusted-relay QKD network. Such performance evaluation method can help us optimize the route in a real-time manner to ensure the fast encryption, decide whether allow newly added services based on the encryption capability of QKD network and monitor the posture of QKD network. The effectiveness of our method is verified by simulation with NS2.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129123933","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 Information Freshness of UAV-assisted IoT Networks Based on DDQN","authors":"Jin Xu, Xiangdong Jia, Zhenchao Hao","doi":"10.1109/IMCEC55388.2022.10019931","DOIUrl":"https://doi.org/10.1109/IMCEC55388.2022.10019931","url":null,"abstract":"In time-sensitive Internet of Things (IoT) systems which is assisted by Unmanned Aerial Vehicle (UA V), most of the existing researches focus on maximizing the system throughput or minimizing the delay, but these researches has ignored the timeliness of the information at the receiver. This paper proposes a UAV-assisted IoT fresh information collection system model, and studies the information collection process and information freshness in the system. In this system, the base station sends the UAV to fly to the sensor node and collect the data in the sensor node. In the process of data collection, the energy of the UAV must be kept in the surplus state. Once the energy of the UAV is lower than the energy threshold, the UAV immediately flies to the destination. The data collected from the sensor node is stored in the buffer and follows the proposed queuing policy which replace old packets with new packets. In this paper, the problem is modeled as a Markov Decision Process and the state space, action space and reward function of the problem are defined. The system average age of information is minimized by jointly optimizing the flight trajectory of UAV and the transmission scheduling sequence of sensors. In order to overcome the disaster of dimension, this paper proposes a node data collection algorithm based on Double Deep Q Learing(DDQN). A large number of simulation experiments show that the proposed DDQN algorithm can reduce the system average age of information effectively compared with other baseline algorithms.","PeriodicalId":423151,"journal":{"name":"2022 IEEE 5th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129261853","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}