{"title":"TBNet: Two-Branch Cloud Detection Network for Remote Sensing Imagery","authors":"Chao Zhao, Huilan Guo, Yanrong Guo, Sheng Zhong, Hangzai Luo, Jianping Fan","doi":"10.1109/ICFEICT57213.2022.00090","DOIUrl":"https://doi.org/10.1109/ICFEICT57213.2022.00090","url":null,"abstract":"In recent years, convolutional neural network-based cloud detection methods have gradually become mainstream and achieved good performance. However, the detection effect of these methods on cloud edge regions and thin cloud regions is generally poor, mainly because these confusing regions account for a small proportion but have problems of greater intra-class inconsistency and stronger inter-class similarity. To this end, we propose a novel two-branch cloud detection network named TBNet, which focuses more on cloud edge regions by adding a boundary prediction branch to the semantic segmentation branch and jointly optimizing parameters. Specifically, the boundary prediction branch is designed to be shallow and pooling-free to preserve details as much as possible while controlling the computational cost of the model. Further-more, Fusion Module and Semantic Enhanced Module are proposed and used for the interaction between the semantic segmentation branch and the boundary prediction branch. FM can provide the necessary high-level features for the boundary prediction branch, and SEM can provide multi-scale contextual information to enable detection of clouds with different characteristics. In the experiments, our method obtains 97.23% accuracy, 93.20% F1-score, 91.46% Kappa, and 91.92% mIoU on the GF-1 wide field-of-view (WFV) satellite imagery dataset.","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114982743","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}
Zehao Ye, Yawei Song, Liangfa Hua, Shengxiang Zhou, Kai Yan, Wei Han
{"title":"Radar Target Tracking Based on Some Multi-Dimensional Adaptive UKF","authors":"Zehao Ye, Yawei Song, Liangfa Hua, Shengxiang Zhou, Kai Yan, Wei Han","doi":"10.1109/ICFEICT57213.2022.00013","DOIUrl":"https://doi.org/10.1109/ICFEICT57213.2022.00013","url":null,"abstract":"When the system model is inaccurate or the noise is uncertain, the traditional UKF algorithm has problems such as decreased filtering accuracy or even divergence. Aiming at these problems, this paper proposes an UKF algorithm based on multi-dimensional adaptive factors (SMA-UKF). Firstly, the sufficient conditions for establishing strong tracking UKF are expounded combined with UKF algorithm and strong tracking filtering principle. Furthermore, some multi-dimensional adaptive factors are introduced into the single-step forward prediction covariance matrix, and their calculation methods are designed respectively. Finally, the effects of SMA-UKF, strong tracking UKF algorithm (ST-UKF) and UKF algorithm in target tracking are simulated and compared under the condition that the system model and noise are inaccurate. The results show that SMA-UKF can automatically judge and adaptively adjust the process noise, and achieve good tracking of the target.","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121034077","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 Wideband Schiffman Phase Shifter Based on SISL Platform","authors":"F. Zhang, Yongqiang Wang, Kaixue Ma","doi":"10.1109/ICFEICT57213.2022.00039","DOIUrl":"https://doi.org/10.1109/ICFEICT57213.2022.00039","url":null,"abstract":"A class of substrate integrated suspended line (SISL) platform-based broadband Schiffman phase shifter is newly proposed. By loading the open-circuited stubs as compensation capacitors, the problem of unequal phase velocity of odd-mode and even-mode existing in parallel coupled-line is solved, which significantly improved the phase shifter's return loss. According to the simulation results, the return loss is better than 27 dB over the operating frequency band, and it reaches 48 dB at the optimal point. A 90° phase shift is basically achieved near the center frequency, and the measured phase imbalance is within ± 5° from 3.27 GHz to 4.72 GHz.","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121364096","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":"Variable Temperature Test and Transient Test of Energetic Materials Based on Perturbation Method","authors":"Yuchong Tang, Yong Gao","doi":"10.1109/ICFEICT57213.2022.00009","DOIUrl":"https://doi.org/10.1109/ICFEICT57213.2022.00009","url":null,"abstract":"In recent years, fireworks production safety accidents have occurred in China, so fireworks and firecrackers safety production testing is an urgent problem to be solved. Based on the above background, this paper proposes a performance test method for energetic materials based on the perturbation method, including two parts of the variable temperature test experiment and the transient detection experiment of the energetic material, of which the variable temperature test experiment shows that the temperature change of the energetic material can be judged based on the perturbation method, and then the transient detection of the explosion of the energetic material is realized by the proposed transient detection method. The temporal resolution of transient detection can reach the microsecond range, which can meet the needs of explosion prediction detection of energetic materials.","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127329788","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}
Tianqi Feng, Yu Shi, E. Li, Hongyu An, Xiongbin Fan, Xingzeng Cha, Dakun Lai
{"title":"Simplifying THz-CT-CW Imaging System Using Vector Network Analyzer and Two Ellipsoid Mirrors","authors":"Tianqi Feng, Yu Shi, E. Li, Hongyu An, Xiongbin Fan, Xingzeng Cha, Dakun Lai","doi":"10.1109/ICFEICT57213.2022.00028","DOIUrl":"https://doi.org/10.1109/ICFEICT57213.2022.00028","url":null,"abstract":"The traditional THz-CT-CW (CT: computed tomography; CW: continuous wave) imaging system separately employs a transmitter and a receiver that extra requires a chopper due to the principle of the receiver. And that system uses at least 4 lenses/mirrors to first collimate the emitted wave and then focus it on the sample. In this paper, we applied a vector network analyzer (VNA) to the THz-CT-CW system to replace the transceiver devices and omit the optical chopper for improving the transceiver device integration. And learning from the free-space method system, we adopt 2 ellipsoid mirrors to the THz-CT-CW system to directly focus the emitted wave on the sample for reducing the number of lenses/mirrors used in the traditional system. Finally, we verified the feasibility of the proposed system according to the imaging results from the experiment. The results show that VNA can be replaceable in THz-CT-CW and using the 2 ellipsoid mirrors is feasible.","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127529139","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}
Daowu Yang, Qi Yang, Xibei Huang, Qiong Liu, Jing Zou
{"title":"3D Point Cloud Feature Information Assisted Audio Source Separation","authors":"Daowu Yang, Qi Yang, Xibei Huang, Qiong Liu, Jing Zou","doi":"10.1109/ICFEICT57213.2022.00069","DOIUrl":"https://doi.org/10.1109/ICFEICT57213.2022.00069","url":null,"abstract":"Separating target singing voices from music vide-os is a challenging task. In this work, we first explore the mono multi-instrument separation problem from a 3D visu-al perspective, combining multimodal features learned from audio and 3D point cloud patterns together. Next, we use 3D sparse convolution to extract sparse tensor features of the point cloud, and use a self-attention mechanism to obtain effective spatial and color information in a series of discrete points. A combination of dense join and dilation convolution unravels the different sound source information in the sound spectrogram signal. The fusion module fuses the features extracted from the visual model and the corresponding feature information obtained from the audio model to finally perform audio source separation. The results show that the 3D point cloud feature representation of musical instruments can better assist audio source separation, which is more applicable to 3D realistic conditions. It also shows that the use of natural synchronization of 3D visual and audio allows the application to better solve problems in 3D virtual or augmented reality applications.","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125961257","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":"Committee: ICFEICT 2022","authors":"","doi":"10.1109/icfeict57213.2022.00006","DOIUrl":"https://doi.org/10.1109/icfeict57213.2022.00006","url":null,"abstract":"","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"131 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120886750","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":"Panoramic Video Streaming Based on Splitting","authors":"Zhuozheng Wang, Qi Xue, Wei Liu","doi":"10.1109/ICFEICT57213.2022.00059","DOIUrl":"https://doi.org/10.1109/ICFEICT57213.2022.00059","url":null,"abstract":"As the development of virtual reality technology growing rapidly, the demand of delivering 360-degree panoramic video stream becomes an important task. High quality panoramic video is often achieved by using high resolution and high framerates and due to these features, delivering high quality panoramic video has considerable data flow, consuming large data bandwidth and computing power. To address this problem, we introduce a novel streaming method, reducing streaming data flow by splitting panoramic video into blocks and selectively deliver the blocks users see in the panoramic video stream, exploiting the fact that the viewer has a limited field of view. Results show that our method effectively reduces the bandwidth when delivering panoramic video without compromising the quality of the panoramic video at the same time.","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114555494","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":"Robust Beamforming for Massive MIMO System with Arbitrary Array Geometry","authors":"Xin Chen, Zhiyong Luo, Congduan Li","doi":"10.1109/ICFEICT57213.2022.00044","DOIUrl":"https://doi.org/10.1109/ICFEICT57213.2022.00044","url":null,"abstract":"Digital beamforming method has been widely applied in MIMO systems, but it is not suitable for massive MIMO systems due to the unbearable power consumptions and costs it causes. Hybrid beamforming becomes a cost-effective solution for massive MIMO beamforming, though there is usually trade-off between performance and computation complexity. How to design a hybrid beamforming method to achieve good performance while with a low complexity becomes a challenge. In this paper, we propose such a method based on eigenvalue decomposition and regularized channel diagonalization (EVD-RCD). Without knowing prior knowl-edge of the array geometry, the proposed algorithm not only achieves high spectral efficiency with low complexity, but also has strong robustness to deal with different antenna array geometry, imperfect CSI and quantized phase shifter.","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116197994","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 Lightweight Anonymous Key Agreement Scheme in V2G Networks","authors":"Huang Chendong, Wang Feng, Huang Zhongming","doi":"10.1109/ICFEICT57213.2022.00062","DOIUrl":"https://doi.org/10.1109/ICFEICT57213.2022.00062","url":null,"abstract":"Electric vehicle access to the grid (V2G) technology enables bidirectional interaction of energy and information between electric vehicles and the grid. The energy transactions between electric vehicles and the power grid is completed by the aggregators. Energy transactions are generally completed through open wireless communication networks, which are vulnerable to network attacks such as eavesdropping, impersonation attack, etc. Data transmission in the V2G networks requires to verify the identity of each communication entity to achieve confidential communication. The realization of confidential communication can enhance the security of energy transactions. Key agreement is a key technique to solve this problem. Therefore, we propose a lightweight anonymous key agreement scheme for V2G networks. For efficiency, we hire a data center to assist electric vehicles and aggregators in our key agreement scheme. Security analysis shows that our scheme can provide anonymity, resist eavesdropping, impersonation and replay attacks, etc. Performance analysis shows that our scheme has lower overall computational cost and communication overhead.","PeriodicalId":175659,"journal":{"name":"2022 2nd International Conference on Frontiers of Electronics, Information and Computation Technologies (ICFEICT)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126449851","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}