2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI)最新文献_第7页

Multi-Person Pose Estimation on Embedded Device 基于嵌入式设备的多人姿态估计

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00020

Zhipeng Ma, Dawei Tian, Ming Zhang, Dingxin He

引用次数: 1

Deep neural network pruning algorithm based on particle swarm 基于粒子群的深度神经网络剪枝算法

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00084

Shengnan Zhang, Shanshan Hong, Chao Wu, Yu Liu, Xiaoming Ju

{"title":"Deep neural network pruning algorithm based on particle swarm","authors":"Shengnan Zhang, Shanshan Hong, Chao Wu, Yu Liu, Xiaoming Ju","doi":"10.1109/ICHCI51889.2020.00084","DOIUrl":"https://doi.org/10.1109/ICHCI51889.2020.00084","url":null,"abstract":"A powerful neural network consumes a lot of storage space and computing resources, which is unacceptable for mobile devices and embedded devices with limited resources. To solve this problem, PSOPruner was proposed. Firstly, The algorithm randomly initializes a series of particles as the pruned network structure; secondly, calculates the threshold corresponding to each convolutional layer according to the particles, deletes the convolution kernels with corresponding values less than the threshold; then, adapts the pruned network structure evaluation, update the particles and the global optimal fitness particles; finally, retrain the network structure corresponding to the global optimal fitness particles to restore the network accuracy. The experimental results show that the algorithm is applied to VGG16 on the cifar-10 data set. After 100 iterations of the algorithm, the test accuracy is improved by 0.04%, the model size is reduced by 94%, and the running speed is increased by 36% compared with the untrimmed model. The pruned network model is more conducive to deployment in mobile devices and embedded devices.","PeriodicalId":355427,"journal":{"name":"2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117007676","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}

引用次数: 0

NBIoT Optimization on massive devices access load control 海量设备接入负载控制的NBIoT优化

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00015

Weinan Cao, Jianzheng Wang, Yifeng Zhao, Lianfeng Huang

{"title":"NBIoT Optimization on massive devices access load control","authors":"Weinan Cao, Jianzheng Wang, Yifeng Zhao, Lianfeng Huang","doi":"10.1109/ICHCI51889.2020.00015","DOIUrl":"https://doi.org/10.1109/ICHCI51889.2020.00015","url":null,"abstract":"Narrow-band Internet of Things (NBIoT) supports a large number of machine connections, random access congestion comes with burst and uncertainty of terminal access. Considering the four key steps of NBIoT random access, this paper models the NBIoT random access process combing with time slot analysis and coverage level transition mechanism. The collision probability and the number of successfully connected devices are derived. Aiming at the fact that the existing ACB algorithm cannot effectively solve the problem of access load control when congestion is severe, this paper proposes a load access control algorithm based on reinforcement learning. In this algorithm, the base station dynamically learns the changes in the congestion state of the system, and adjusts the access level restriction parameters accordingly to reduce the collision probability. The simulation results show that the proposed algorithm system can quickly converge, effectively reduce the probability of access collisions under congestion conditions, increase the access success rate, and improve system access performance.","PeriodicalId":355427,"journal":{"name":"2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI)","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132752957","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}

引用次数: 1

Review of Text analysis Based on Deep Learning 基于深度学习的文本分析综述

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00087

Liu Ying, Li Huidi

引用次数: 3

Design of New Media Data Processing System 新媒体数据处理系统设计

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00059

Hui Shan, Hongshen Zhang, Xueming Mi

引用次数: 0

Multi-Model Inference Acceleration on Embedded Multi-Core Processors 嵌入式多核处理器的多模型推理加速

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00090

Peiqi Shi, Feng Gao, Songtao Liang, Shanjin Yu

{"title":"Multi-Model Inference Acceleration on Embedded Multi-Core Processors","authors":"Peiqi Shi, Feng Gao, Songtao Liang, Shanjin Yu","doi":"10.1109/ICHCI51889.2020.00090","DOIUrl":"https://doi.org/10.1109/ICHCI51889.2020.00090","url":null,"abstract":"The predominant resource efficient approaches that enable on-device inference include designing lightweight DNN architectures like MobileNets, SqueezeNets, compressing model using techniques such as network pruning, vector quantization, distillation, binarization. Recent research on using dynamic layer-wise partitioning and partial execution of CNN based model inference also make it possible to co-inference on memory and computation resource constrained devices. However, these approaches have their own bottleneck, lightweight DNN architectures and model compression usually compromise accuracy in order to deploy on resource constrained devices, dynamic model partitioning efficiency depends heavily on the network condition. This paper proposes an approach for multimodel inference acceleration on heterogeneous devices. The idea is to deploy multiple single object detection model instead of one heavy multiple object, this is because in most cases it only needs to detect one or two objects in one scenario and single object detection model weight could be lighter for the same resolution quality and require less resource. Moreover, in cloud-edge-device scenario, with the help of a scheduler policy, it is possible to gradually update models in need.","PeriodicalId":355427,"journal":{"name":"2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115721409","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}

引用次数: 1

Scheduling Optimization of real-time IOT system based on RNN 基于RNN的实时物联网系统调度优化

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00061

Shenling Liu, Chunyuan Zhang, Yujiao Chen

{"title":"Scheduling Optimization of real-time IOT system based on RNN","authors":"Shenling Liu, Chunyuan Zhang, Yujiao Chen","doi":"10.1109/ICHCI51889.2020.00061","DOIUrl":"https://doi.org/10.1109/ICHCI51889.2020.00061","url":null,"abstract":"Ubiquitous computation, which promoted by Rapid development of Wireless Sensor Network (WSN) technologies cuts across many areas of modern day living, Internet of Things has been identified as one of Network Infrastructure in the next generation of application domains. An architecture based on cloud computing at the center, which contribute to highly flexibility and scalablity, is an extensively used scheme to construct IOT applications. With growing number of intelligent terminals and third part application accessing on the platform, the Qos problem caused by the large-scale concurrent access rise to the surface. To address this question, a self-adaption Multi-level Feedback Queue Scheduling policy, used to reduce mean turnaround time and complexity of scheduling, based on Recurrent Neural Network (RNN) is presented in this paper. Feature parameters of queues and tasks are used as input of network, the calculated parameters are exported to optimize queue parameters continuously. This research implement a prototype of this scheme. According To demonstrate the efficiency, this thesis give performance results from our prototype and other scheduling policy.","PeriodicalId":355427,"journal":{"name":"2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116303951","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}

引用次数: 0

A school bullying detecting algorithm based on motion recognition and speech emotion recognition 一种基于动作识别和语音情感识别的校园欺凌检测算法

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00066

C. Wei, Hua Zhang, Liang Ye, Fanchao Meng

{"title":"A school bullying detecting algorithm based on motion recognition and speech emotion recognition","authors":"C. Wei, Hua Zhang, Liang Ye, Fanchao Meng","doi":"10.1109/ICHCI51889.2020.00066","DOIUrl":"https://doi.org/10.1109/ICHCI51889.2020.00066","url":null,"abstract":"School bullying is a common social problem among teenagers. It affects the victims both mentally and physically, and is considered as one of the main reasons for depression, dropping out of school, and adolescent suicide. For this reason, preventing school bullying is significant to the student’s mental and physical health. In order to detect bullying events in time, this paper proposes a bullying detecting algorithm based on motion recognition and speech emotion recognition. People wear an electronic equipment, which is used to collect his/her motion and speech data, to detect bullying events in real-time. In this paper, the authors extract five features from acceleration and gyro data for physical bullying detection. The PLP features are extracted for verbal bullying detection. Then authors use the Relief-F algorithm for feature selection, and the PPCA algorithm is used to reduce the dimensionality of the feature matrix. Finally, the authors use the KNN algorithm as the classifier to train the motion recognition model and the SVM algorithm as the classifier to train the speech emotion recognition model. With cross-validation, the average accuracy of the motion recognition system is 80.61%, whereas that of the speech emotion recognition system is 75.76%. The simulation results of the algorithm indicate that the anti-bullying detecting algorithm could identify the bullying event effectively.","PeriodicalId":355427,"journal":{"name":"2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116526043","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}

引用次数: 2

Processing Optical Fiber Sensing Signals with Compressed Sensing under special working conditions 特殊工况下压缩感知处理光纤传感信号

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00035

Liu Hang, Wang Bo, Lang Daizhi, Huang Rongqiang

引用次数: 0

Defect Detection Algorithm of Anti-vibration Hammer Based on Improved Cascade R-CNN 基于改进级联R-CNN的减振锤缺陷检测算法

2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI) Pub Date : 2020-12-01 DOI: 10.1109/ICHCI51889.2020.00070

Bao Wenxia, Ren Yangxun, Liangyu Dong, Yang Xianjun, Xu Qiuju

{"title":"Defect Detection Algorithm of Anti-vibration Hammer Based on Improved Cascade R-CNN","authors":"Bao Wenxia, Ren Yangxun, Liangyu Dong, Yang Xianjun, Xu Qiuju","doi":"10.1109/ICHCI51889.2020.00070","DOIUrl":"https://doi.org/10.1109/ICHCI51889.2020.00070","url":null,"abstract":"Aiming at the problem that it is difficult to accurately locate and identify the defects of anti-vibration hammer components in high-voltage transmission lines, this paper propose a detection method for anti-vibration hammer defects based on the improved Cascade R-CNN algorithm. In dataset: Firstly, this research construct a dataset of anti-vibration hammer defects based on common anti-vibration hammer defect categories; secondly, this research perform preprocessing methods such as cropping, flipping, gamma transformation and CLAHE on training samples to improve the generalization ability of the network and avoid over-fitting. In algorithm: This research use ResNeXt-101 as the backbone network of the Cascade R-CNN algorithm; add FPN module for extracting multi-scale features to extract more effective information; use Focal Loss function to improve the classification loss of RPN module to solve the dataset category imbalance problem. Experimental results show that the improved Cascade R-CNN algorithm has a detection accuracy of 91.2% on the anti-vibration hammer defect test set, which is 3.5% higher than the original Cascade R-CNN algorithm and is better than other mainstream object detection algorithms.","PeriodicalId":355427,"journal":{"name":"2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124984764","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}

引用次数: 5