Pub Date : 2019-11-01DOI: 10.1109/IICSPI48186.2019.9095874
Zheng Zhao, Guoqing Liu, Dewen Li
Aiming at the status of the measurement of the deposition dust thickness of ventilation and dust collection pipes in industrial workplaces in China, the law of dust distribution in ventilation and dust collection pipelines was studied by numerical simulation method. A high-precision dust deposition thickness detection method was studied. For this reason, based on the gas-dust flow characteristics in the ventilation and dust removal pipelines, based on the SIMPLE algorithm of the same-location grid, the dust distribution law of the ventilation and dust removal pipelines was numerically simulated using Fluent; then the dust deposition quality-thickness relationship was derived using the simulation results. The expression, which converts quality inspection into thickness inspection, completes this detection method. Experiments show that: the thickness of the detection of the resolution of 0.01mm and the accuracy of 0.07mm achieved good detection results.
{"title":"Deposition Thickness Detection Method based on Dust Distribution Law of Ventilation Dust Removal Pipeline","authors":"Zheng Zhao, Guoqing Liu, Dewen Li","doi":"10.1109/IICSPI48186.2019.9095874","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9095874","url":null,"abstract":"Aiming at the status of the measurement of the deposition dust thickness of ventilation and dust collection pipes in industrial workplaces in China, the law of dust distribution in ventilation and dust collection pipelines was studied by numerical simulation method. A high-precision dust deposition thickness detection method was studied. For this reason, based on the gas-dust flow characteristics in the ventilation and dust removal pipelines, based on the SIMPLE algorithm of the same-location grid, the dust distribution law of the ventilation and dust removal pipelines was numerically simulated using Fluent; then the dust deposition quality-thickness relationship was derived using the simulation results. The expression, which converts quality inspection into thickness inspection, completes this detection method. Experiments show that: the thickness of the detection of the resolution of 0.01mm and the accuracy of 0.07mm achieved good detection results.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133231487","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}
To improve the safety of UAV operating in airspace, TOPAZ method is introduced for the safety assessment of UAV flight conflict resolution technology. Firstly, the conflict scenes including $45^{circ}, 90^{circ}$ crossing encounter for single-aircraft and simultaneous crossing encounter for double-aircraft are added to construct the airspace operation environment for UAV. Then a typical UAV conflict resolution algorithm (a modified GA algorithm) is selected as the safety assessment object and is systematically evaluated according to the process of TOPAZ. Finally, a dynamic heuristic factor is introduced to modify the selected algorithm and validate the improvement of safety level of the algorithm by TOPAZ.
{"title":"Improvement of Safety Assessment by UsingTOPAZ Process for UAV Conflict Resolution","authors":"Xusheng Gan, Baohua Zhang, Xiangwei Meng, Liying Ding","doi":"10.1109/IICSPI48186.2019.9095930","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9095930","url":null,"abstract":"To improve the safety of UAV operating in airspace, TOPAZ method is introduced for the safety assessment of UAV flight conflict resolution technology. Firstly, the conflict scenes including $45^{circ}, 90^{circ}$ crossing encounter for single-aircraft and simultaneous crossing encounter for double-aircraft are added to construct the airspace operation environment for UAV. Then a typical UAV conflict resolution algorithm (a modified GA algorithm) is selected as the safety assessment object and is systematically evaluated according to the process of TOPAZ. Finally, a dynamic heuristic factor is introduced to modify the selected algorithm and validate the improvement of safety level of the algorithm by TOPAZ.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"127 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123584394","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 : 2019-11-01DOI: 10.1109/IICSPI48186.2019.9095895
Shuren Zhou, Jia Qiu
This paper designs a module of attention regions in SSD detector for accurate and efficient object detection (RSSD). Different from previous one-stage detection method like SSD which just simply applied the multi-scale head-features and directly extracted from backbone network, for classification and regression, our method aims to strengthen the characterization of head-features further. The parallel encode-to-decode structure is constructed and a computation method of regional distribution on features (R-Softmax) is proposed. What’s more, in order to reduce time-costs, the down-sampling layers are shared with the multi-scale layers from backbone network. Our detector performs better on PASCAL VOC datasets (e.g., 78.4% mAP V.S. SSD 76.4% on VOC 07test) and costs 0.001s per image more than SSD.
{"title":"RSSD: Object Detection via Attention Regions in SSD Detector","authors":"Shuren Zhou, Jia Qiu","doi":"10.1109/IICSPI48186.2019.9095895","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9095895","url":null,"abstract":"This paper designs a module of attention regions in SSD detector for accurate and efficient object detection (RSSD). Different from previous one-stage detection method like SSD which just simply applied the multi-scale head-features and directly extracted from backbone network, for classification and regression, our method aims to strengthen the characterization of head-features further. The parallel encode-to-decode structure is constructed and a computation method of regional distribution on features (R-Softmax) is proposed. What’s more, in order to reduce time-costs, the down-sampling layers are shared with the multi-scale layers from backbone network. Our detector performs better on PASCAL VOC datasets (e.g., 78.4% mAP V.S. SSD 76.4% on VOC 07test) and costs 0.001s per image more than SSD.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122744171","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 : 2019-11-01DOI: 10.1109/IICSPI48186.2019.9095877
Song Li, S. Ning, Yao Yezhou, Tian Jingjing, Zhang Wenxue, Chi Liang
According to multisource quality safety data of defective automobile products, key quality safety factors of defective automobile products are extracted, a defect information indicator system for automobile products is systematically constructed and a correlated graph is established between quality safety factors. Based on the optimization and correlation of the quality safety factor indicator system, Big Data technology is used to design a data structure for multisource quality safety information cluster, develop a data platform for the defect information analysis of automobile products and achieve information clustering and correlation analysis based on multisource quality safety data, providing technical support for the recall management of defective automobile products.
{"title":"Application of Data Mining Technology in the Recall of Defective Automobile Products in China ——A Typical Case of the Construction of Digital China","authors":"Song Li, S. Ning, Yao Yezhou, Tian Jingjing, Zhang Wenxue, Chi Liang","doi":"10.1109/IICSPI48186.2019.9095877","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9095877","url":null,"abstract":"According to multisource quality safety data of defective automobile products, key quality safety factors of defective automobile products are extracted, a defect information indicator system for automobile products is systematically constructed and a correlated graph is established between quality safety factors. Based on the optimization and correlation of the quality safety factor indicator system, Big Data technology is used to design a data structure for multisource quality safety information cluster, develop a data platform for the defect information analysis of automobile products and achieve information clustering and correlation analysis based on multisource quality safety data, providing technical support for the recall management of defective automobile products.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124792730","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 : 2019-11-01DOI: 10.1109/IICSPI48186.2019.9096049
Jiawei Wang, Hongyun Xiong
Recent work has shown that deep convolutional neural networks have made immensely successful in many computer vision tasks such as semantic image segmentation, which can be thought as a complexed localization and classification problem. However, due to the limitation of computing cost and memory, most existing models are difficult to deploy on mobile devices. It is also an arduous task to get more semantic information from the feature map of downsampling. In this paper, we introduce cascaded low-rank convolutions network (CLRCNet) which is an efficient neural network by using multiple low-rank layers. The cascaded low-rank layers are used to reduce computational complexity. A pooling operation in network units is introduced to learn more contextual information during training. A large number of experiments show that the method has better performance than other network structures. Our network struct attains mean intersection over union (mIOU) of 63.3% on Cityscapes dataset at 76.9 frames per second on $512 times 1024$ resolution.
{"title":"CLRCNet: Cascaded Low-rank Convolutions for Semantic Segmentation in Real-time","authors":"Jiawei Wang, Hongyun Xiong","doi":"10.1109/IICSPI48186.2019.9096049","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9096049","url":null,"abstract":"Recent work has shown that deep convolutional neural networks have made immensely successful in many computer vision tasks such as semantic image segmentation, which can be thought as a complexed localization and classification problem. However, due to the limitation of computing cost and memory, most existing models are difficult to deploy on mobile devices. It is also an arduous task to get more semantic information from the feature map of downsampling. In this paper, we introduce cascaded low-rank convolutions network (CLRCNet) which is an efficient neural network by using multiple low-rank layers. The cascaded low-rank layers are used to reduce computational complexity. A pooling operation in network units is introduced to learn more contextual information during training. A large number of experiments show that the method has better performance than other network structures. Our network struct attains mean intersection over union (mIOU) of 63.3% on Cityscapes dataset at 76.9 frames per second on $512 times 1024$ resolution.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128297596","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 : 2019-11-01DOI: 10.1109/IICSPI48186.2019.9095926
Lei Yang, Liwei Tian, Longqing Zhang
The freshman registration rate in colleges has always been a concern of all colleges, and the prediction of the number of freshmen before registration is a difficult problem. As a subject originating from artificial intelligence and statistics, machine learning is one of the key research directions in the field of data analysis. At present, no researchers have used machine learning to predict the registration of freshmen, because whether freshmen register or not is a very subjective matter, which is affected by many subjective factors. Nowadays, the traditional methods of predicting the number of new students in Colleges and universities are telephone inquiry and tuition fee information inquiry. According to the data of enrollment and registration in a college in the past years, we use machine learning method to analyze it. The results show that whether freshmen register or not is predictable, and the data of enrollment and registration in colleges in the past years is valuable.
{"title":"Research on Freshman Registration Prediction Based on Machine Learning","authors":"Lei Yang, Liwei Tian, Longqing Zhang","doi":"10.1109/IICSPI48186.2019.9095926","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9095926","url":null,"abstract":"The freshman registration rate in colleges has always been a concern of all colleges, and the prediction of the number of freshmen before registration is a difficult problem. As a subject originating from artificial intelligence and statistics, machine learning is one of the key research directions in the field of data analysis. At present, no researchers have used machine learning to predict the registration of freshmen, because whether freshmen register or not is a very subjective matter, which is affected by many subjective factors. Nowadays, the traditional methods of predicting the number of new students in Colleges and universities are telephone inquiry and tuition fee information inquiry. According to the data of enrollment and registration in a college in the past years, we use machine learning method to analyze it. The results show that whether freshmen register or not is predictable, and the data of enrollment and registration in colleges in the past years is valuable.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126119281","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 : 2019-11-01DOI: 10.1109/IICSPI48186.2019.9095974
Liu Weihao, Chen Jiamin, Wang Ning, Shen Jun, Li Weijiao, Ji Linhua, Chen Xiaodong
With the rapid development of the domestic logistics industry, the demand for quick inquiry of express parcel delivery information is becoming more and more urgent. The automatic acquisition of express delivery number is expected to solve this problem. This paper proposes a barcode localization segmentation recognition algorithm for bar code/QR code location segmentation recognition in a single scan image of a parcel in a complex environment. The whole algorithm is fully tested in the actual express single scan image. The results show that the algorithm is fast, accurate and has low bit error rate, and has strong practical value.
{"title":"Fast segmentation identification of express parcel barcode based on MSRCR enhanced high noise environment","authors":"Liu Weihao, Chen Jiamin, Wang Ning, Shen Jun, Li Weijiao, Ji Linhua, Chen Xiaodong","doi":"10.1109/IICSPI48186.2019.9095974","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9095974","url":null,"abstract":"With the rapid development of the domestic logistics industry, the demand for quick inquiry of express parcel delivery information is becoming more and more urgent. The automatic acquisition of express delivery number is expected to solve this problem. This paper proposes a barcode localization segmentation recognition algorithm for bar code/QR code location segmentation recognition in a single scan image of a parcel in a complex environment. The whole algorithm is fully tested in the actual express single scan image. The results show that the algorithm is fast, accurate and has low bit error rate, and has strong practical value.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126195406","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 : 2019-11-01DOI: 10.1109/IICSPI48186.2019.9095928
Hua Jiang, Qingrui Wang, Jinpo Fan, Gang Zhang
The service chain, which is not dependent on the special hardware facilities and the network topology is changeable, is studied, and an authenticated group key management scheme suitable for service chain is proposed. The scheme is based on the bilinear mapping cryptosystem and combines the threshold idea with the identity authentication method, which improves the efficiency and security of the protocol. This scheme also realizes the connection security between the virtual network functions in the service chain while carrying out group key updating, and proves its correctness and security. The analysis results show that the scheme is suitable for the dynamic key management of service chain with the advantages of small number of wheels and small computing overhead in ensuring the safety of each instance in the service chain.
{"title":"Authenticated Group Key Management Scheme in Service Chain","authors":"Hua Jiang, Qingrui Wang, Jinpo Fan, Gang Zhang","doi":"10.1109/IICSPI48186.2019.9095928","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9095928","url":null,"abstract":"The service chain, which is not dependent on the special hardware facilities and the network topology is changeable, is studied, and an authenticated group key management scheme suitable for service chain is proposed. The scheme is based on the bilinear mapping cryptosystem and combines the threshold idea with the identity authentication method, which improves the efficiency and security of the protocol. This scheme also realizes the connection security between the virtual network functions in the service chain while carrying out group key updating, and proves its correctness and security. The analysis results show that the scheme is suitable for the dynamic key management of service chain with the advantages of small number of wheels and small computing overhead in ensuring the safety of each instance in the service chain.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125435319","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 : 2019-11-01DOI: 10.1109/IICSPI48186.2019.9096019
Dong Xu, Jie Tan
Scleral blood vessels are very stable biological features. Lightweight head-mounted scleral vascular imaging system can be widely used in personal identification, gaze tracking and many other fields. However, it is difficult to acquire clear images of scleral blood vessels at a small distance with traditional optical imaging systems. We proposed a new scleral vascular imaging system based on micro-lens array, which can capture the light field of scleral blood vessels near eyes with sub-image array. The system has a simple and integrative structure. And it can easily reconstruct the 3D position and structure of scleral blood vessels from multiple sub-aperture images.
{"title":"Design of Close Scleral Vascular Imaging System Used for Gazing Tracking","authors":"Dong Xu, Jie Tan","doi":"10.1109/IICSPI48186.2019.9096019","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9096019","url":null,"abstract":"Scleral blood vessels are very stable biological features. Lightweight head-mounted scleral vascular imaging system can be widely used in personal identification, gaze tracking and many other fields. However, it is difficult to acquire clear images of scleral blood vessels at a small distance with traditional optical imaging systems. We proposed a new scleral vascular imaging system based on micro-lens array, which can capture the light field of scleral blood vessels near eyes with sub-image array. The system has a simple and integrative structure. And it can easily reconstruct the 3D position and structure of scleral blood vessels from multiple sub-aperture images.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127646107","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 : 2019-11-01DOI: 10.1109/IICSPI48186.2019.9095900
Yongxiang Feng, Xiaoxin Wang, Leixiao Li
With the rapid development of information technology and internet finance, RFM model technology has been widely used in banking financial services. The traditional financial services industry has gradually changed from product center to customer center. As customer transaction data continues to grow in the database, it is urgent to analyze the transaction information with the high efficiency big data analysis technology. Therefore, an RFM model suitable for financial customers has been established, and this model has been applied to the financial product recommendation guidance system. It can greatly improve the quality of customer marketing services in the banking industry, and can effectively reduce operating costs.
{"title":"The Application Research of Customer Segmentation Model in Bank Financial Marketing","authors":"Yongxiang Feng, Xiaoxin Wang, Leixiao Li","doi":"10.1109/IICSPI48186.2019.9095900","DOIUrl":"https://doi.org/10.1109/IICSPI48186.2019.9095900","url":null,"abstract":"With the rapid development of information technology and internet finance, RFM model technology has been widely used in banking financial services. The traditional financial services industry has gradually changed from product center to customer center. As customer transaction data continues to grow in the database, it is urgent to analyze the transaction information with the high efficiency big data analysis technology. Therefore, an RFM model suitable for financial customers has been established, and this model has been applied to the financial product recommendation guidance system. It can greatly improve the quality of customer marketing services in the banking industry, and can effectively reduce operating costs.","PeriodicalId":318693,"journal":{"name":"2019 2nd International Conference on Safety Produce Informatization (IICSPI)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127739884","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：