Pub Date : 2020-03-01DOI: 10.1109/CTISC49998.2020.00027
Yun Jiang, Shengxin Tao, Hai Zhang, Simin Cao
Deep neural networks usually contain tens to hundreds of millions of orders of learning parameters that provide the necessary representation to solve various visual tasks. But with the increase of the representational ability, the possibility of over-fitting also increase, which bring about poor generalization. In this paper, we propose MA (Maximum Activation point processing) algorithm, a new image data augmentation method which is designed to improve the generalization ability of the model and reduce the risk of overfitting. During the training process, the most discriminative part of the input image is searched for, and the model is driven to search for the supplementary information of the most important feature information by erasing the maximum attention image block. During this process, training images with different occlusion levels are generated as new inputs to the network and the model continues to be trained. The image erasure method based on the maximum activation point guidance only needs to modify the input image, which can effectively improve the robustness of the model to occluded image recognition, and can be integrated with various network structures. The effectiveness of our method is verified on the Cifar10, Cifar100 and Fashion-MNIST datasets.
深度神经网络通常包含数千万到数亿阶的学习参数，这些参数为解决各种视觉任务提供了必要的表示。但随着表征能力的提高，过度拟合的可能性也随之增加，导致泛化效果较差。本文提出了一种新的图像数据增强方法MA (Maximum Activation point processing，最大激活点处理)算法，该算法旨在提高模型的泛化能力，降低过拟合的风险。在训练过程中，搜索输入图像中最具判别性的部分，并通过擦除最大关注图像块来驱动模型搜索最重要特征信息的补充信息。在此过程中，生成不同遮挡水平的训练图像作为网络的新输入，并继续训练模型。基于最大激活点制导的图像擦除方法只需要修改输入图像，可以有效提高模型对遮挡图像识别的鲁棒性，并且可以与各种网络结构集成。在Cifar10、Cifar100和Fashion-MNIST数据集上验证了该方法的有效性。
{"title":"Image data augmentation method based on maximum activation point guided erasure","authors":"Yun Jiang, Shengxin Tao, Hai Zhang, Simin Cao","doi":"10.1109/CTISC49998.2020.00027","DOIUrl":"https://doi.org/10.1109/CTISC49998.2020.00027","url":null,"abstract":"Deep neural networks usually contain tens to hundreds of millions of orders of learning parameters that provide the necessary representation to solve various visual tasks. But with the increase of the representational ability, the possibility of over-fitting also increase, which bring about poor generalization. In this paper, we propose MA (Maximum Activation point processing) algorithm, a new image data augmentation method which is designed to improve the generalization ability of the model and reduce the risk of overfitting. During the training process, the most discriminative part of the input image is searched for, and the model is driven to search for the supplementary information of the most important feature information by erasing the maximum attention image block. During this process, training images with different occlusion levels are generated as new inputs to the network and the model continues to be trained. The image erasure method based on the maximum activation point guidance only needs to modify the input image, which can effectively improve the robustness of the model to occluded image recognition, and can be integrated with various network structures. The effectiveness of our method is verified on the Cifar10, Cifar100 and Fashion-MNIST datasets.","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115293182","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 : 2020-03-01DOI: 10.1109/ctisc49998.2020.00008
{"title":"Sponsors and Supporters: CTISC 2020","authors":"","doi":"10.1109/ctisc49998.2020.00008","DOIUrl":"https://doi.org/10.1109/ctisc49998.2020.00008","url":null,"abstract":"","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"467 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115368436","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 : 2020-03-01DOI: 10.1109/ctisc49998.2020.00036
{"title":"CTISC 2020 Index","authors":"","doi":"10.1109/ctisc49998.2020.00036","DOIUrl":"https://doi.org/10.1109/ctisc49998.2020.00036","url":null,"abstract":"","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121660957","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 : 2020-03-01DOI: 10.1109/ctisc49998.2020.00005
We hope to continue our communication by sharing new research ideas, discussing challenges, and forming collaboration to solve various issues on Computer Science. Despite the influence of the pandemic, CTISC 2020 provides a flexible platform for the delegates to exchange new ideas and application experiences face to face, to establish business or research relations and to find global partners for future collaboration. Experts, scholars, and practitioners in the field of computer Technology, Information Science and Communications have the opportunity to exchange new ideas and present state-of-the-art researches.
{"title":"CTISC 2020 Opinion","authors":"","doi":"10.1109/ctisc49998.2020.00005","DOIUrl":"https://doi.org/10.1109/ctisc49998.2020.00005","url":null,"abstract":"We hope to continue our communication by sharing new research ideas, discussing challenges, and forming collaboration to solve various issues on Computer Science. Despite the influence of the pandemic, CTISC 2020 provides a flexible platform for the delegates to exchange new ideas and application experiences face to face, to establish business or research relations and to find global partners for future collaboration. Experts, scholars, and practitioners in the field of computer Technology, Information Science and Communications have the opportunity to exchange new ideas and present state-of-the-art researches.","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122457515","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 : 2020-03-01DOI: 10.1109/CTISC49998.2020.00021
Xiujuan Xie, Xiangju Li
The blended learning mode is the inevitable trend of the teaching reform in Colleges and universities, and the new teaching mode inevitably needs a matching assessment mechanism. Therefore, this paper explores the teaching design and assessment mechanism under the blended learning mode, taking the program design course of application-oriented universities as an example. Guided by the ability, reconstruct teaching content, and carry out online and offline blended learning through three stages of “before class, during class and after class”. Then construct a multi angle, multi form and multi-agent course process assessment system. After a semester of teaching practice, compared with previous years, students’ enthusiasm and initiative in learning have been improved, classroom participation and course attention have been significantly improved, and the teaching effect and learning efficiency of the course have been improved.
{"title":"Exploration and Practice of Process Assessment and Evaluation Method Based on Blended Learning : ——Take programming courses as an example","authors":"Xiujuan Xie, Xiangju Li","doi":"10.1109/CTISC49998.2020.00021","DOIUrl":"https://doi.org/10.1109/CTISC49998.2020.00021","url":null,"abstract":"The blended learning mode is the inevitable trend of the teaching reform in Colleges and universities, and the new teaching mode inevitably needs a matching assessment mechanism. Therefore, this paper explores the teaching design and assessment mechanism under the blended learning mode, taking the program design course of application-oriented universities as an example. Guided by the ability, reconstruct teaching content, and carry out online and offline blended learning through three stages of “before class, during class and after class”. Then construct a multi angle, multi form and multi-agent course process assessment system. After a semester of teaching practice, compared with previous years, students’ enthusiasm and initiative in learning have been improved, classroom participation and course attention have been significantly improved, and the teaching effect and learning efficiency of the course have been improved.","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124597530","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 : 2020-03-01DOI: 10.1109/CTISC49998.2020.00020
Xiaoming Zhu, Baisen Liu, Yan Wang, Xiaoguang Wang
UWB is a radio engineering technology with ultra wide spectrum resource, which is mainly applied in the short-range high speed communication, image transmission and wireless sensor network. A compact planar monopole antenna with size 18.4×32.4×1.6mm3 is proposed for terminals of portable UWB mobile systems. The wider impedance matching characteristic of proposed antenna is obtained by adjusting antenna structure based on transmission line model theory. The new resonant frequency points are added by transforming rectangular radiation patch and CPW feeder of original antenna. Finally experimental results show the working bandwidth of proposed antenna is from 2.72GHz to 14.2GHz, which extends six times more than original antenna.
{"title":"Planar Monopole UWB Antenna Design Base on Transmission Line Model","authors":"Xiaoming Zhu, Baisen Liu, Yan Wang, Xiaoguang Wang","doi":"10.1109/CTISC49998.2020.00020","DOIUrl":"https://doi.org/10.1109/CTISC49998.2020.00020","url":null,"abstract":"UWB is a radio engineering technology with ultra wide spectrum resource, which is mainly applied in the short-range high speed communication, image transmission and wireless sensor network. A compact planar monopole antenna with size 18.4×32.4×1.6mm3 is proposed for terminals of portable UWB mobile systems. The wider impedance matching characteristic of proposed antenna is obtained by adjusting antenna structure based on transmission line model theory. The new resonant frequency points are added by transforming rectangular radiation patch and CPW feeder of original antenna. Finally experimental results show the working bandwidth of proposed antenna is from 2.72GHz to 14.2GHz, which extends six times more than original antenna.","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"211 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116524637","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 : 2020-03-01DOI: 10.1109/CTISC49998.2020.00017
Min Wu, Changqing Li
Based on the engineering big data measured by Huawei Mobility Base Station and the application of electronic map, this paper establishes the radio wave feature engineering by combining the Cognitive Radio (CR) and ResNet. Based on the statistical results of channel large-scale fading characteristics, including path loss (PL), shadow fading (SF) and small-scale fading characteristics, the actual Reference Signal Receiving Power (RSRP) of the channel is obtained to modify the traditional empirical model formula of wireless channel, and the model is used to accurately predict the wireless signal coverage in the new environment, so as to greatly reduce the cost of base station construction and improve the network construction efficiency.
{"title":"5G Wireless Intelligent Propagation Channel Modelling Based on Deep Residual Network","authors":"Min Wu, Changqing Li","doi":"10.1109/CTISC49998.2020.00017","DOIUrl":"https://doi.org/10.1109/CTISC49998.2020.00017","url":null,"abstract":"Based on the engineering big data measured by Huawei Mobility Base Station and the application of electronic map, this paper establishes the radio wave feature engineering by combining the Cognitive Radio (CR) and ResNet. Based on the statistical results of channel large-scale fading characteristics, including path loss (PL), shadow fading (SF) and small-scale fading characteristics, the actual Reference Signal Receiving Power (RSRP) of the channel is obtained to modify the traditional empirical model formula of wireless channel, and the model is used to accurately predict the wireless signal coverage in the new environment, so as to greatly reduce the cost of base station construction and improve the network construction efficiency.","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128869354","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 : 2020-03-01DOI: 10.1109/CTISC49998.2020.00034
Jie Cheng, Dong He, Changhee Lee
With benefits from wide-range detection and accurate measurements, LiDAR has been widely studied in the field of autonomous driving. Yet many challenges remain in modern LiDAR 3D point clouds processing algorithms. One is ground segmentation due to the real-time requirement dealing with huge input data from LiDAR. In this study, we propose a method to separate ground points from a point cloud in a hybrid way by adopting dynamic section division, height-based conditional filter, and multi-lines linear regression. Where, physical characteristics of LiDAR mounted on a vehicle have been introduced in the dynamic section division. Thereafter, we raise a conditional filter algorithm for filtering outliers of point clouds, and use multi-lines linear regression to generate the ground skeleton. In the end, the qualitative and quantitative experiments validate the performance by using two datasets and indicate that our proposed method outperforms state-of-the-art methods on KITTI in terms of accuracy 94.1% and runtime 432ms. The source code is publicly available under https://github.com/0-0cj/sgs
{"title":"A simple ground segmentation method for LiDAR 3D point clouds","authors":"Jie Cheng, Dong He, Changhee Lee","doi":"10.1109/CTISC49998.2020.00034","DOIUrl":"https://doi.org/10.1109/CTISC49998.2020.00034","url":null,"abstract":"With benefits from wide-range detection and accurate measurements, LiDAR has been widely studied in the field of autonomous driving. Yet many challenges remain in modern LiDAR 3D point clouds processing algorithms. One is ground segmentation due to the real-time requirement dealing with huge input data from LiDAR. In this study, we propose a method to separate ground points from a point cloud in a hybrid way by adopting dynamic section division, height-based conditional filter, and multi-lines linear regression. Where, physical characteristics of LiDAR mounted on a vehicle have been introduced in the dynamic section division. Thereafter, we raise a conditional filter algorithm for filtering outliers of point clouds, and use multi-lines linear regression to generate the ground skeleton. In the end, the qualitative and quantitative experiments validate the performance by using two datasets and indicate that our proposed method outperforms state-of-the-art methods on KITTI in terms of accuracy 94.1% and runtime 432ms. The source code is publicly available under https://github.com/0-0cj/sgs","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"157 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116427951","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 : 2020-03-01DOI: 10.1109/CTISC49998.2020.00012
Nanzhou Hu, Shanghui Xiao, S. Shao
Analog self-interference(SI) cancellation in full-duplex systems can be implemented with either fixed or adjustable delay arrangement. In this paper, the theoretical residual SI power in multi-tap analog SI cancellation with arbitrary time delay arrangement for each tap is derived, and the cancellation performance with fixed and optimal adjustable delay arrangement are compared by simulations. The influence of the bandwidth, the carrier frequency, the number of taps and the maximum delay of the SI channel on the cancellation performance are investigated. In consideration of the complexity of the optimal method, a novel algorithm for adjustable delay arrangement is proposed based on the cross-correlation of the reference signal and the SI signal.
{"title":"Analysis on Delay Arrangement of Analog Self-Interference Cancellation for Full-Duplex","authors":"Nanzhou Hu, Shanghui Xiao, S. Shao","doi":"10.1109/CTISC49998.2020.00012","DOIUrl":"https://doi.org/10.1109/CTISC49998.2020.00012","url":null,"abstract":"Analog self-interference(SI) cancellation in full-duplex systems can be implemented with either fixed or adjustable delay arrangement. In this paper, the theoretical residual SI power in multi-tap analog SI cancellation with arbitrary time delay arrangement for each tap is derived, and the cancellation performance with fixed and optimal adjustable delay arrangement are compared by simulations. The influence of the bandwidth, the carrier frequency, the number of taps and the maximum delay of the SI channel on the cancellation performance are investigated. In consideration of the complexity of the optimal method, a novel algorithm for adjustable delay arrangement is proposed based on the cross-correlation of the reference signal and the SI signal.","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125327816","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 : 2020-03-01DOI: 10.1109/CTISC49998.2020.00022
Wenbing Fan, Zhenzheng Zhang
In order to improve the efficiency and prediction accuracy of wastewater treatment in the printing and dyeing industry, in view of the difficulty of measuring the content of wastewater indicator BOD (Biochemical Oxygen Demand), this paper proposes a Convolutional Neural Network and Support Vector Regression hybrid model of wastewater index content prediction model. Firstly, the input easy-to-measure data is constructed in sequence in the form of a window as a model input. Secondly, CNN is used to extract feature vectors. The resulting feature vectors are constructed in a sequence and used as input data for SVR. Finally, SVR is used for index prediction, and compare with convolutional neural network model and support vector regression model. The actual wastewater treatment plant data in the UCI database is used for experiments. The mean absolute error (MAE) and root mean squared error (RMSE) are used as the evaluation criteria. The experimental results show that the CNN-SVR hybrid model proposed in this paper with higher prediction accuracy.
{"title":"A CNN-SVR Hybrid Prediction Model for Wastewater Index Measurement","authors":"Wenbing Fan, Zhenzheng Zhang","doi":"10.1109/CTISC49998.2020.00022","DOIUrl":"https://doi.org/10.1109/CTISC49998.2020.00022","url":null,"abstract":"In order to improve the efficiency and prediction accuracy of wastewater treatment in the printing and dyeing industry, in view of the difficulty of measuring the content of wastewater indicator BOD (Biochemical Oxygen Demand), this paper proposes a Convolutional Neural Network and Support Vector Regression hybrid model of wastewater index content prediction model. Firstly, the input easy-to-measure data is constructed in sequence in the form of a window as a model input. Secondly, CNN is used to extract feature vectors. The resulting feature vectors are constructed in a sequence and used as input data for SVR. Finally, SVR is used for index prediction, and compare with convolutional neural network model and support vector regression model. The actual wastewater treatment plant data in the UCI database is used for experiments. The mean absolute error (MAE) and root mean squared error (RMSE) are used as the evaluation criteria. The experimental results show that the CNN-SVR hybrid model proposed in this paper with higher prediction accuracy.","PeriodicalId":266384,"journal":{"name":"2020 2nd International Conference on Advances in Computer Technology, Information Science and Communications (CTISC)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133381187","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：