Pub Date : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874115
Wentao Zhang, Kun Liu, Heng Li
Aiming at the defects of poor identification effect and prone to be disturbed by weather and illumination changes in vehicle detection using a single sensor, a multi-sensor fusion sensing system based on millimeter wave radar and camera was designed in this paper. Firstly, the spatial and temporal coordinates of millimeter wave radar and camera are unified through coordinate transformation and time alignment. Then, YOLOV5 deep neural network model is used to realize target detection of camera data, including cars, trucks and buses. Finally, data fusion is realized according to the detection results of the two sensors. Through field experiments, the vehicle detection accuracy reaches 95.3%. The results show that the proposed system overcomes the deficiency of single sensor in target detection, which can improve the reliability and effectiveness of vehicle detection.
{"title":"Traffic vehicle detection by fusion of millimeter wave radar and camera","authors":"Wentao Zhang, Kun Liu, Heng Li","doi":"10.1109/ISPDS56360.2022.9874115","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874115","url":null,"abstract":"Aiming at the defects of poor identification effect and prone to be disturbed by weather and illumination changes in vehicle detection using a single sensor, a multi-sensor fusion sensing system based on millimeter wave radar and camera was designed in this paper. Firstly, the spatial and temporal coordinates of millimeter wave radar and camera are unified through coordinate transformation and time alignment. Then, YOLOV5 deep neural network model is used to realize target detection of camera data, including cars, trucks and buses. Finally, data fusion is realized according to the detection results of the two sensors. Through field experiments, the vehicle detection accuracy reaches 95.3%. The results show that the proposed system overcomes the deficiency of single sensor in target detection, which can improve the reliability and effectiveness of vehicle detection.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122763479","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 : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874110
Lisha Chen, Jiawei Liu
In view of the large deviation of landscape pattern information extraction results caused by many types of landscape patterns and strong interference factors, a landscape pattern information extraction method based on Airborne Hyperspectral Images is proposed. Relevant images are collected through the imaging and interpretation process of ground object spectra, and the remote sensing images are decomposed and processed. The decomposed images are fused by Laplace method. On this basis, according to the second-order neighborhood difference algorithm of Markov random field model, the energy function in the background is extracted, the non target landscape pattern information is suppressed, and the target area of landscape pattern is calibrated. The spectral vector is added in front of the projection operator, and the background and landscape pattern information of the calibration area are separated by means of low probability detection algorithm to realize the extraction of landscape pattern information. The experimental results show that the proposed method has high integrity, short running time and high accuracy.
{"title":"Extraction of landscape pattern information from Airborne Hyperspectral Images","authors":"Lisha Chen, Jiawei Liu","doi":"10.1109/ISPDS56360.2022.9874110","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874110","url":null,"abstract":"In view of the large deviation of landscape pattern information extraction results caused by many types of landscape patterns and strong interference factors, a landscape pattern information extraction method based on Airborne Hyperspectral Images is proposed. Relevant images are collected through the imaging and interpretation process of ground object spectra, and the remote sensing images are decomposed and processed. The decomposed images are fused by Laplace method. On this basis, according to the second-order neighborhood difference algorithm of Markov random field model, the energy function in the background is extracted, the non target landscape pattern information is suppressed, and the target area of landscape pattern is calibrated. The spectral vector is added in front of the projection operator, and the background and landscape pattern information of the calibration area are separated by means of low probability detection algorithm to realize the extraction of landscape pattern information. The experimental results show that the proposed method has high integrity, short running time and high accuracy.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"132 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131657977","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 : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874143
Yu Zhang, Jingbo Peng, Xiaobo Zhang, Yang Yu, Fei Zhang, Hao Wang
With a focus on the problem of long time delay of aero-engine networked control system, a robust controller is designed in this paper. The model of the aero-engine networked control system considering modelling error and external disturbances is first established. Then a robust controller is designed based on Lyapunov stability theory and matrix inequality methods. To verified the effectiveness of the proposed method, numerical simulations are made and the results reflect that the proposed controller can effectively stabilize the system in a short time despite the existence of uncertainties.
{"title":"Robust Control of Aero-engine Networked Control System with Long Time Delay","authors":"Yu Zhang, Jingbo Peng, Xiaobo Zhang, Yang Yu, Fei Zhang, Hao Wang","doi":"10.1109/ISPDS56360.2022.9874143","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874143","url":null,"abstract":"With a focus on the problem of long time delay of aero-engine networked control system, a robust controller is designed in this paper. The model of the aero-engine networked control system considering modelling error and external disturbances is first established. Then a robust controller is designed based on Lyapunov stability theory and matrix inequality methods. To verified the effectiveness of the proposed method, numerical simulations are made and the results reflect that the proposed controller can effectively stabilize the system in a short time despite the existence of uncertainties.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"232 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129068740","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 : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874066
Si Wei, Hui Xi, Kaiwang Zhang, Yijia Yun, Haoran Li
To enhance prediction reliability and accuracy, an Lstm model optimized by the improved grey wolf algorithm is introduced for daily air quality index forecasting. Firstly, the model preprocesses the collected data and divides the data into a training set and a testing set. Then, using Tent Chaotic Sequence to generate an initial population, which increases the diversity of individuals in the population; And aming at the shortage of the search ability of Grey Wolf Optimization (GWO), updating the parameters $a$. The improved GWO (IGWO) used to optimize the relevant hyperparameters in the long and short-term memory neural network. Finally, the IGWO-LSTM model constructed with excellent hyperparameters will use the test set to obtain the prediction results. The experimental results demonstrate the proposed method outperforms the other four model in AQI prediction.
{"title":"Air Quality Time Series Prediction Optimized by Grey Wolf Algorithm","authors":"Si Wei, Hui Xi, Kaiwang Zhang, Yijia Yun, Haoran Li","doi":"10.1109/ISPDS56360.2022.9874066","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874066","url":null,"abstract":"To enhance prediction reliability and accuracy, an Lstm model optimized by the improved grey wolf algorithm is introduced for daily air quality index forecasting. Firstly, the model preprocesses the collected data and divides the data into a training set and a testing set. Then, using Tent Chaotic Sequence to generate an initial population, which increases the diversity of individuals in the population; And aming at the shortage of the search ability of Grey Wolf Optimization (GWO), updating the parameters $a$. The improved GWO (IGWO) used to optimize the relevant hyperparameters in the long and short-term memory neural network. Finally, the IGWO-LSTM model constructed with excellent hyperparameters will use the test set to obtain the prediction results. The experimental results demonstrate the proposed method outperforms the other four model in AQI prediction.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117160379","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 : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874053
Jiahao Zeng, Decheng Wang, Peng Chen
Aiming at the problems that the traditional color region growing segmentation algorithm has a large amount of computation, slow running speed and is easily affected by noise, this paper proposes an improved color region growing point cloud segmentation algorithm based on octree. The proposed algorithm consists of two segmentation stages from coarse to fine: firstly, an octree-based voxelized representation of the input point cloud is performed, and a traditional region growing algorithm segmentation step is performed to extract the main (coarse) parts. Then, the region growth of boundary points is performed by replacing geometric features with color features to achieve fine segmentation. The experimental results show that this method can not only effectively segment point cloud data, but also solve the problem of instability of traditional color-based region growth segmentation, and improve the accuracy, reliability and running speed of point cloud segmentation.
{"title":"Improved color region growing point cloud segmentation algorithm based on octree","authors":"Jiahao Zeng, Decheng Wang, Peng Chen","doi":"10.1109/ISPDS56360.2022.9874053","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874053","url":null,"abstract":"Aiming at the problems that the traditional color region growing segmentation algorithm has a large amount of computation, slow running speed and is easily affected by noise, this paper proposes an improved color region growing point cloud segmentation algorithm based on octree. The proposed algorithm consists of two segmentation stages from coarse to fine: firstly, an octree-based voxelized representation of the input point cloud is performed, and a traditional region growing algorithm segmentation step is performed to extract the main (coarse) parts. Then, the region growth of boundary points is performed by replacing geometric features with color features to achieve fine segmentation. The experimental results show that this method can not only effectively segment point cloud data, but also solve the problem of instability of traditional color-based region growth segmentation, and improve the accuracy, reliability and running speed of point cloud segmentation.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125611857","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}
It is extremely dangerous for epilepsy patients to become sick when no one is accompanying them in their daily life. The alarm for epilepsy patients can be timely notified to their families to take measures. In this context, a scheme for the identification of general tonic-clonic seizures (GTCs) based on wrist signals is proposed. Firstly, features were extracted from wrist acceleration(ACC), skin conductance response(SCR), number of wrist movements(NOWM) and heart rate(HR) signals. Secondly, in order to reduce the interference of unnecessary features on classification, feature dimensions were reduced by random forest algorithm. Finally, the number of normal data samples is much larger than the number of diseased data samples, and the training model is adopted to sacrifice the accuracy of identifying diseased data and improve the accuracy of identifying normal data. The detection and recognition effects of SVM (Support vector machine), AdaBoost and XGBoost machine learning models are compared. The results showed that the SVM algorithm could recognize all GTCs episodes (median 39.5s, range 5-69s) in the 10 data with a false recognition rate (FRR) of 0.08/d when the continuous predicted onset time reached 9s. When the predicted onset time reaches 19s, the three algorithm models can effectively reduce FRR, but at the same time, more underreporting will be generated. GTCs seizures can be detected through wrist signals, and it has good recognition effect and low FRR, which is conducive to the development of wearable epilepsy recognition devices.
{"title":"Total tonic clonic seizure recognition of wrist signals","authors":"Guangliang Xu, Chang Chen, Jing Wang, Yi'nan Zhou, Tingwei Liang","doi":"10.1109/ISPDS56360.2022.9874082","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874082","url":null,"abstract":"It is extremely dangerous for epilepsy patients to become sick when no one is accompanying them in their daily life. The alarm for epilepsy patients can be timely notified to their families to take measures. In this context, a scheme for the identification of general tonic-clonic seizures (GTCs) based on wrist signals is proposed. Firstly, features were extracted from wrist acceleration(ACC), skin conductance response(SCR), number of wrist movements(NOWM) and heart rate(HR) signals. Secondly, in order to reduce the interference of unnecessary features on classification, feature dimensions were reduced by random forest algorithm. Finally, the number of normal data samples is much larger than the number of diseased data samples, and the training model is adopted to sacrifice the accuracy of identifying diseased data and improve the accuracy of identifying normal data. The detection and recognition effects of SVM (Support vector machine), AdaBoost and XGBoost machine learning models are compared. The results showed that the SVM algorithm could recognize all GTCs episodes (median 39.5s, range 5-69s) in the 10 data with a false recognition rate (FRR) of 0.08/d when the continuous predicted onset time reached 9s. When the predicted onset time reaches 19s, the three algorithm models can effectively reduce FRR, but at the same time, more underreporting will be generated. GTCs seizures can be detected through wrist signals, and it has good recognition effect and low FRR, which is conducive to the development of wearable epilepsy recognition devices.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120985613","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 : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874232
Jialin Yu, Jun Liang, Haoyang Mei, Jingwen Fan, Songsen Yu
Few-shot learning is an approach that classify unseen classes with limited labeled samples. We propose improved networks of Relation Network to classify images with small samples. The improved networks is ECA Relation Network (ECA-RNET). The accuracy of ECA-RNET is 52.24% and 67.85% on 5-way 1-shot and 5-way 5-shot of mini-ImageNet dataset, respectively.
{"title":"Improved Few-Shot Learning for Images Classification","authors":"Jialin Yu, Jun Liang, Haoyang Mei, Jingwen Fan, Songsen Yu","doi":"10.1109/ISPDS56360.2022.9874232","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874232","url":null,"abstract":"Few-shot learning is an approach that classify unseen classes with limited labeled samples. We propose improved networks of Relation Network to classify images with small samples. The improved networks is ECA Relation Network (ECA-RNET). The accuracy of ECA-RNET is 52.24% and 67.85% on 5-way 1-shot and 5-way 5-shot of mini-ImageNet dataset, respectively.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128899399","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 : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874150
Rao Fu, Jiandong Fang, Yudong Zhao
In order to recognize the daily behaviors of cattle in an open environment, the daily behaviors of cattle were classified based on the image features in the dynamic region. Firstly, the target detection model was used to locate the cattle feature parts in the dynamic region of the image, and the image features in the dynamic region were extracted according to the label information of the feature parts, then, the deep neural network was used to classify the image features. Finally, the results show that in the open environment, the accuracy of the model in predicting the feeding, lying and standing behaviors of cattle was 84%.
{"title":"Daily behavior recognition of cattle based on dynamic region image features in open environment","authors":"Rao Fu, Jiandong Fang, Yudong Zhao","doi":"10.1109/ISPDS56360.2022.9874150","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874150","url":null,"abstract":"In order to recognize the daily behaviors of cattle in an open environment, the daily behaviors of cattle were classified based on the image features in the dynamic region. Firstly, the target detection model was used to locate the cattle feature parts in the dynamic region of the image, and the image features in the dynamic region were extracted according to the label information of the feature parts, then, the deep neural network was used to classify the image features. Finally, the results show that in the open environment, the accuracy of the model in predicting the feeding, lying and standing behaviors of cattle was 84%.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133944413","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 : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874175
Weiguo Yi, Siwei Ma, Heng Zhang, B. Ma
Absrtact: Aiming at the problem of clothing classification, a convolution neural network based on vgg16 is proposed. Firstly, the color and name data of clothing are labeled, and then trained on vgg16 model; Finally, vgg16 model is fine tuned and added to migration learning. The results show that the accuracy of this method is higher than that of the original model, which is suitable for garment classification and has a good application prospect.
{"title":"Classification and improvement of multi label image based on vgg16 network","authors":"Weiguo Yi, Siwei Ma, Heng Zhang, B. Ma","doi":"10.1109/ISPDS56360.2022.9874175","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874175","url":null,"abstract":"Absrtact: Aiming at the problem of clothing classification, a convolution neural network based on vgg16 is proposed. Firstly, the color and name data of clothing are labeled, and then trained on vgg16 model; Finally, vgg16 model is fine tuned and added to migration learning. The results show that the accuracy of this method is higher than that of the original model, which is suitable for garment classification and has a good application prospect.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133365700","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 : 2022-07-22DOI: 10.1109/ISPDS56360.2022.9874023
Anqi Xu, Wenhao Liu, Dibin Zhou
With the rapid development and wide application of flexible circuit board (FPC) industry, the intelligent demand of flexible circuit board wiring design has become increasingly urgent. This paper mainly studies FPC automatic wiring algorithm based on line optimization strategy. Using line optimization and maze addressing method, it has the advantages of high efficiency, low cost and high degree of automation. It can provide a good solution to the problem of long cycle, low efficiency, time-consuming and laborious of traditional keyboard wiring.
{"title":"Research on Algorithm of Intelligent Keyboard Routing Based on Line Priority Strategy","authors":"Anqi Xu, Wenhao Liu, Dibin Zhou","doi":"10.1109/ISPDS56360.2022.9874023","DOIUrl":"https://doi.org/10.1109/ISPDS56360.2022.9874023","url":null,"abstract":"With the rapid development and wide application of flexible circuit board (FPC) industry, the intelligent demand of flexible circuit board wiring design has become increasingly urgent. This paper mainly studies FPC automatic wiring algorithm based on line optimization strategy. Using line optimization and maze addressing method, it has the advantages of high efficiency, low cost and high degree of automation. It can provide a good solution to the problem of long cycle, low efficiency, time-consuming and laborious of traditional keyboard wiring.","PeriodicalId":280244,"journal":{"name":"2022 3rd International Conference on Information Science, Parallel and Distributed Systems (ISPDS)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132932930","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：