Pub Date : 2023-01-06DOI: 10.1109/ICCECE58074.2023.10135197
Zhili Liu, Heyang Sun, Jinliang Song, Bin Zhang, Yuhang Yan, Bingbing Qiu, Lihang Jiang, Jingjing Li
With the development of the electric power system, the construction of electric power credit has achieved positive results, but there is still a certain gap compared with the requirements of the government and enterprises. In this paper, a vertical federated learning framework including user credit evaluation is proposed. By constructing a vertical federated learning credit sharing system between electric power companies and financial companies, the information barriers of both are reduced and the market transaction risks are reduced. Through the construction of refined electricity price pricing model based on user credit evaluation, it is beneficial to reduce the cost and increase the efficiency of users, and encourage users to develop with high credit and high quality.
{"title":"Vertical Federated Learning Architecture for Power Company and Financial Company and Electricity Pricing Model Considering User Credit Evaluation","authors":"Zhili Liu, Heyang Sun, Jinliang Song, Bin Zhang, Yuhang Yan, Bingbing Qiu, Lihang Jiang, Jingjing Li","doi":"10.1109/ICCECE58074.2023.10135197","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135197","url":null,"abstract":"With the development of the electric power system, the construction of electric power credit has achieved positive results, but there is still a certain gap compared with the requirements of the government and enterprises. In this paper, a vertical federated learning framework including user credit evaluation is proposed. By constructing a vertical federated learning credit sharing system between electric power companies and financial companies, the information barriers of both are reduced and the market transaction risks are reduced. Through the construction of refined electricity price pricing model based on user credit evaluation, it is beneficial to reduce the cost and increase the efficiency of users, and encourage users to develop with high credit and high quality.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131559186","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}
Pub Date : 2023-01-06DOI: 10.1109/ICCECE58074.2023.10135281
Jiabin Zhang, Xiaoru Wang, Han Xia, Xiaolong Li
Recently, several computer vision tasks have begun to adopt transformer-based approaches with promising results. Using a completely transformer-based architecture in image recovery achieves better performance than the existing CNN approach, but the existing vision transformers lack the scalability for high-resolution images, which means that transformers are underutilized in image restoration tasks. We propose a hybrid architecture (HCT) that uses both CNN and transformer to improve image restoration. HCT consists of transformer and CNN branches. By fully integrating the two branches, we strengthen the network's ability of parameter sharing and local information aggregation, and also increase the network's ability to integrate global information, and finally achieve the purpose of improving the image recovery effect. Our proposed transformer branch uses a spatial fusion adaptive attention model that blends local and global attention improving image restoration while reducing computing costs. Extensive experiments show that HCT achieves competitive results in super-resolution tasks.
{"title":"HCT: Hybrid CNN-Transformer Networks for Super-Resolution","authors":"Jiabin Zhang, Xiaoru Wang, Han Xia, Xiaolong Li","doi":"10.1109/ICCECE58074.2023.10135281","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135281","url":null,"abstract":"Recently, several computer vision tasks have begun to adopt transformer-based approaches with promising results. Using a completely transformer-based architecture in image recovery achieves better performance than the existing CNN approach, but the existing vision transformers lack the scalability for high-resolution images, which means that transformers are underutilized in image restoration tasks. We propose a hybrid architecture (HCT) that uses both CNN and transformer to improve image restoration. HCT consists of transformer and CNN branches. By fully integrating the two branches, we strengthen the network's ability of parameter sharing and local information aggregation, and also increase the network's ability to integrate global information, and finally achieve the purpose of improving the image recovery effect. Our proposed transformer branch uses a spatial fusion adaptive attention model that blends local and global attention improving image restoration while reducing computing costs. Extensive experiments show that HCT achieves competitive results in super-resolution tasks.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127696149","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}
Pub Date : 2023-01-06DOI: 10.1109/ICCECE58074.2023.10135325
Weijuan Li, R. Jia
We present SA-MVSNet, a novel two-stage multi-view stereo network equipped with self-attention mechanism, which can improve the quality of low-resolution image 3D reconstruction. SA-MVSNet consists of two stages, and the lower resolution depth maps predicted in the first stage provide a priori information for the second stage. To increase the utilization of image information, a pyramid scheme was used to fuse the feature maps at different resolutions. Moreover, we introduce an improved self-attention module in the first stage to improve reconstruction accuracy by learning the long-term dependence information of feature maps. The experiments on the DTU dataset show a promising result in both completeness and accuracy metrics of the 3D scene reconstructed by the proposed method.
{"title":"A Self-Attention based Network for Low Resolution Multi-View Stereo","authors":"Weijuan Li, R. Jia","doi":"10.1109/ICCECE58074.2023.10135325","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135325","url":null,"abstract":"We present SA-MVSNet, a novel two-stage multi-view stereo network equipped with self-attention mechanism, which can improve the quality of low-resolution image 3D reconstruction. SA-MVSNet consists of two stages, and the lower resolution depth maps predicted in the first stage provide a priori information for the second stage. To increase the utilization of image information, a pyramid scheme was used to fuse the feature maps at different resolutions. Moreover, we introduce an improved self-attention module in the first stage to improve reconstruction accuracy by learning the long-term dependence information of feature maps. The experiments on the DTU dataset show a promising result in both completeness and accuracy metrics of the 3D scene reconstructed by the proposed method.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132629244","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}
Pub Date : 2023-01-06DOI: 10.1109/ICCECE58074.2023.10135369
Heng Zhang, Lishan Jia
ATC training simulation system is widely used in controller training. The track display module is an important part of ATC training simulation system. The low-cost simulation system based on microcomputer has the characteristics of low cost and high simulation degree. This paper is based on the modeling method of improved Euler angle formula, and then improves the longitude and latitude update algorithm. Finally, the GL Studio graphic designer updates the control interface design according to the standard instrument approach diagram of Capital Airport, and uses the virtual pilot model to realize the simulation of the track display module in the aircraft approach phase through the VC++software compilation platform. The practice proves that the design method makes the simulation system interface clear and the aircraft target motion real-time.
{"title":"Construction of Aircraft Approach Simulation System based on Virtual Pilot Model","authors":"Heng Zhang, Lishan Jia","doi":"10.1109/ICCECE58074.2023.10135369","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135369","url":null,"abstract":"ATC training simulation system is widely used in controller training. The track display module is an important part of ATC training simulation system. The low-cost simulation system based on microcomputer has the characteristics of low cost and high simulation degree. This paper is based on the modeling method of improved Euler angle formula, and then improves the longitude and latitude update algorithm. Finally, the GL Studio graphic designer updates the control interface design according to the standard instrument approach diagram of Capital Airport, and uses the virtual pilot model to realize the simulation of the track display module in the aircraft approach phase through the VC++software compilation platform. The practice proves that the design method makes the simulation system interface clear and the aircraft target motion real-time.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132240409","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}
Pub Date : 2023-01-06DOI: 10.1109/ICCECE58074.2023.10135538
Sicong Chen, Chaobing Huang
The use of remote sensing images for ship detection can accurately monitor ship targets and provide reliable reference for monitoring key sea areas. Since the horizontal detection model cannot precisely locate and represent the specific direction of the ship, we propose a rotation detector based on YOLOv5m6 and KFIoU, which can realize the detection of ships in arbitrary orientations. On the other hand, the punishment based on Gaussian Wasserstein distance is used in model to generate confidence loss, which improves the discrimination between foreground and background during ship detection. Finally, transformer pyramid attention is added to the backbone of network, which uses the fusion of information extracted in multi-scale space and the self-attention mechanism to improve the feature extraction effect and the accuracy of detection. On FGSD2021 dataset, our model finally achieves 88.24% of mAP after adding attention mechanism and improving the confidence loss.
{"title":"RDYOLOv5m6-KF: A Rotation Detector for Ship Detection in Remote Sensing Images","authors":"Sicong Chen, Chaobing Huang","doi":"10.1109/ICCECE58074.2023.10135538","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135538","url":null,"abstract":"The use of remote sensing images for ship detection can accurately monitor ship targets and provide reliable reference for monitoring key sea areas. Since the horizontal detection model cannot precisely locate and represent the specific direction of the ship, we propose a rotation detector based on YOLOv5m6 and KFIoU, which can realize the detection of ships in arbitrary orientations. On the other hand, the punishment based on Gaussian Wasserstein distance is used in model to generate confidence loss, which improves the discrimination between foreground and background during ship detection. Finally, transformer pyramid attention is added to the backbone of network, which uses the fusion of information extracted in multi-scale space and the self-attention mechanism to improve the feature extraction effect and the accuracy of detection. On FGSD2021 dataset, our model finally achieves 88.24% of mAP after adding attention mechanism and improving the confidence loss.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134155305","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}
Pub Date : 2023-01-06DOI: 10.1109/ICCECE58074.2023.10135328
Song Ren, Meng Ding
Recently, works related to video action recognition focus on using hybrid streams as input to get better results. Those streams usually are combinations of RGB channel with one additional feature stream such as audio, optical flow and pose information. Among those extra streams, posture as unstructured data is more difficult to fuse with RGB channel than the others. In this paper, we propose our Heavy Pose Empowered RGB Nets (HPER-Nets) ‐‐an end-to-end multitasking model‐‐based on the thorough investigation on how to fuse posture and RGB information. Given video frames as the only input, our model will reinforce it by merging the intrinsic posture information in the form of part affinity fields (PAFs), and use this hybrid stream to perform further video action recognition. Experimental results show that our model can outperform other different methods on UCF-101, UMDB and Kinetics datasets, and with only 16 frames, a 95.3% Top-1 accuracy on UCF101, a 69.6% on HMDB and a 41.0% on Kinetics have been recorded.
{"title":"Heavy Pose Empowered RGB Nets for Video Action Recognition","authors":"Song Ren, Meng Ding","doi":"10.1109/ICCECE58074.2023.10135328","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135328","url":null,"abstract":"Recently, works related to video action recognition focus on using hybrid streams as input to get better results. Those streams usually are combinations of RGB channel with one additional feature stream such as audio, optical flow and pose information. Among those extra streams, posture as unstructured data is more difficult to fuse with RGB channel than the others. In this paper, we propose our Heavy Pose Empowered RGB Nets (HPER-Nets) ‐‐an end-to-end multitasking model‐‐based on the thorough investigation on how to fuse posture and RGB information. Given video frames as the only input, our model will reinforce it by merging the intrinsic posture information in the form of part affinity fields (PAFs), and use this hybrid stream to perform further video action recognition. Experimental results show that our model can outperform other different methods on UCF-101, UMDB and Kinetics datasets, and with only 16 frames, a 95.3% Top-1 accuracy on UCF101, a 69.6% on HMDB and a 41.0% on Kinetics have been recorded.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115882792","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}
Pub Date : 2023-01-06DOI: 10.1109/ICCECE58074.2023.10135263
Zexuan Chen, Lan Wang
In view of the actual need to predict aircraft trajectory, traditional prediction models often have problems such as insufficient precision and slow training efficiency. By analyzing the target trajectory with temporal characteristics, the Elastic-BiGRU trajectory prediction model is proposed, which combines the Smooth filtering method, the Elastic Network fitting method and the GRU structure, the prediction accuracy of aircraft trajectory is further improved. The experimental results show that the Elastic-BiGRU model compared with Bi-LSTM model and Bi-GRU model, its MSE error is relatively reduced by more than 8% and 11%The Elastic-BiGRU also solves the problem of slow training speed of Bi-LSTM model, and saves about 20% of the time.
{"title":"Aircraft Trajectory Prediction Model Based on Improved GRU Structure","authors":"Zexuan Chen, Lan Wang","doi":"10.1109/ICCECE58074.2023.10135263","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135263","url":null,"abstract":"In view of the actual need to predict aircraft trajectory, traditional prediction models often have problems such as insufficient precision and slow training efficiency. By analyzing the target trajectory with temporal characteristics, the Elastic-BiGRU trajectory prediction model is proposed, which combines the Smooth filtering method, the Elastic Network fitting method and the GRU structure, the prediction accuracy of aircraft trajectory is further improved. The experimental results show that the Elastic-BiGRU model compared with Bi-LSTM model and Bi-GRU model, its MSE error is relatively reduced by more than 8% and 11%The Elastic-BiGRU also solves the problem of slow training speed of Bi-LSTM model, and saves about 20% of the time.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115922987","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}
Pub Date : 2023-01-06DOI: 10.1109/ICCECE58074.2023.10135338
Jianing Li, Yunfei Zhu
This paper presents a model based on a 3-layer feedforward neural network, which effectively preserves the characteristics of the chemical content of each category in ancient glass through 3 fully connected layers. The average prediction rate of the model was 96.43%, which was 2.43% higher than the traditional KNN classification model, 3.42% higher than the support vector machine (SVM) model and 8.43% higher than the random forest model, demonstrating the efficiency of the model.
{"title":"Composition analysis and identification of ancient glass objects based on neural network models","authors":"Jianing Li, Yunfei Zhu","doi":"10.1109/ICCECE58074.2023.10135338","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135338","url":null,"abstract":"This paper presents a model based on a 3-layer feedforward neural network, which effectively preserves the characteristics of the chemical content of each category in ancient glass through 3 fully connected layers. The average prediction rate of the model was 96.43%, which was 2.43% higher than the traditional KNN classification model, 3.42% higher than the support vector machine (SVM) model and 8.43% higher than the random forest model, demonstrating the efficiency of the model.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114985914","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}
Pub Date : 2023-01-06DOI: 10.1109/ICCECE58074.2023.10135354
Ziliang Liu, Hongwe Chen
The Harris Hawk optimization (HHO) algorithm is an excellent swarm intelligence optimization algorithm which has the advantages of high efficiency in finding the best, ease of implementation and wide application. It also has some disadvantages such as the possibility of convergence too fast and the tendency to fall into local optima. This paper combines an improved escape energy update approach and the leader update operator of the Salp Swarm Algorithm to improve the HHO, named IMHHO. The experiments show that the improvements have improved the algorithm's ability to find the best. IMHHO was also used in the parameter optimization of the Extreme Learning Machine, which also enables the ELM to find the right weights and bias values and to regress the data more accurately.
{"title":"An improved Harris Hawk optimization algorithm and its application to Extreme Learning Machine","authors":"Ziliang Liu, Hongwe Chen","doi":"10.1109/ICCECE58074.2023.10135354","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135354","url":null,"abstract":"The Harris Hawk optimization (HHO) algorithm is an excellent swarm intelligence optimization algorithm which has the advantages of high efficiency in finding the best, ease of implementation and wide application. It also has some disadvantages such as the possibility of convergence too fast and the tendency to fall into local optima. This paper combines an improved escape energy update approach and the leader update operator of the Salp Swarm Algorithm to improve the HHO, named IMHHO. The experiments show that the improvements have improved the algorithm's ability to find the best. IMHHO was also used in the parameter optimization of the Extreme Learning Machine, which also enables the ELM to find the right weights and bias values and to regress the data more accurately.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115533040","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 normal operation of power transformer is related to the safety and stability of the power grid. Abnormal temperature may cause damage to transformer equipment, seriously affect its service life, and even lead to major accidents. In this paper, a transformer temperature prediction method based on big data is proposed. The ambient temperature is included in the prediction conditions. A feature extraction method based on adaptive weighting is designed to mine the time series features in the column head temperature and ambient temperature, and an interactive feature fusion strategy is used to form a comprehensive and reliable transformer temperature prediction. The experimental simulation shows that the transformer temperature prediction method proposed in this paper has high prediction accuracy, effectively provides more quantitative auxiliary information for the operation monitoring of power transformer equipment, ensures the safe and stable operation of transformer, and has high practicability.
{"title":"Big Data Analysis Based Transformer Temperature Prediction Method in Distribution Station Area","authors":"Xianming Cheng, Haipeng Sun, Zhibin Yin, Xiao Ding","doi":"10.1109/ICCECE58074.2023.10135457","DOIUrl":"https://doi.org/10.1109/ICCECE58074.2023.10135457","url":null,"abstract":"The normal operation of power transformer is related to the safety and stability of the power grid. Abnormal temperature may cause damage to transformer equipment, seriously affect its service life, and even lead to major accidents. In this paper, a transformer temperature prediction method based on big data is proposed. The ambient temperature is included in the prediction conditions. A feature extraction method based on adaptive weighting is designed to mine the time series features in the column head temperature and ambient temperature, and an interactive feature fusion strategy is used to form a comprehensive and reliable transformer temperature prediction. The experimental simulation shows that the transformer temperature prediction method proposed in this paper has high prediction accuracy, effectively provides more quantitative auxiliary information for the operation monitoring of power transformer equipment, ensures the safe and stable operation of transformer, and has high practicability.","PeriodicalId":120030,"journal":{"name":"2023 3rd International Conference on Consumer Electronics and Computer Engineering (ICCECE)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124761408","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：