Pub Date : 2022-05-01DOI: 10.1109/cniot55862.2022.00034
Shuzhen Chen, Pinghui Chen, Hai Huang, Zu-Xin Wu, Minwei Yang, Nuoya Zhang, Lei Wang
5G end-to-end network slice is a great enabler to provide customization, dedication and isolation of network resources to meet different industries/users special requirements. 5G UE network slicing based on UE Route Selection Policy (URSP) plays an important role in realizing end-to-end network slice. However, there are some key issues required to be resolved in 5G UE, like traffic descriptor transferring and authentication. Some current potential solutions do not meet these key issues. In this paper, we put forward and implement a 5G UE network slicing based on SDK to resolve these key issues. And we do have the end-to-end network slicing filed test for the 5G UE network slicing SDK. Our field test results confirm the successful usage of end-to-end 5G network slicing by proposed 5G UE network slicing SDK solution.
{"title":"5G UE Network Slicing based on SDK Implementation and Field Test","authors":"Shuzhen Chen, Pinghui Chen, Hai Huang, Zu-Xin Wu, Minwei Yang, Nuoya Zhang, Lei Wang","doi":"10.1109/cniot55862.2022.00034","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00034","url":null,"abstract":"5G end-to-end network slice is a great enabler to provide customization, dedication and isolation of network resources to meet different industries/users special requirements. 5G UE network slicing based on UE Route Selection Policy (URSP) plays an important role in realizing end-to-end network slice. However, there are some key issues required to be resolved in 5G UE, like traffic descriptor transferring and authentication. Some current potential solutions do not meet these key issues. In this paper, we put forward and implement a 5G UE network slicing based on SDK to resolve these key issues. And we do have the end-to-end network slicing filed test for the 5G UE network slicing SDK. Our field test results confirm the successful usage of end-to-end 5G network slicing by proposed 5G UE network slicing SDK solution.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"153 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122186647","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00033
Hanxiao Xue, Wenfei Wan, Shengyun Wei
The Just Noticeable Difference (JND) model involves the minimum level of visibility where visual content can be distinguished, which plays a significant role in terms of the perceptual image/video. For the estimation of the JND threshold, the contrast masking effect evaluation is a critical task and has room for manoeuvre. Considering the important role of structural information for contrast masking evaluation, a structure complexity descriptor based on orientation selectivity characteristics of the human visual system (HVS) was introduced and a new contrast masking model based on structure complexity in the discrete cosine transformation (DCT) domain was estimated. Then, combining with the spatio-temporal CSF and luminance adaptation, a novel JND model for videos was proposed in the DCT domain. The experimental results of the subjective quality evaluation tests demonstrate that the proposed JND threshold can hide more noises under the same perceived quality, which is highly consistent with human subjective visual perception.
{"title":"A Novel DCT-based Just Noticeable Difference Model for Videos Based on Structure Complexity","authors":"Hanxiao Xue, Wenfei Wan, Shengyun Wei","doi":"10.1109/cniot55862.2022.00033","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00033","url":null,"abstract":"The Just Noticeable Difference (JND) model involves the minimum level of visibility where visual content can be distinguished, which plays a significant role in terms of the perceptual image/video. For the estimation of the JND threshold, the contrast masking effect evaluation is a critical task and has room for manoeuvre. Considering the important role of structural information for contrast masking evaluation, a structure complexity descriptor based on orientation selectivity characteristics of the human visual system (HVS) was introduced and a new contrast masking model based on structure complexity in the discrete cosine transformation (DCT) domain was estimated. Then, combining with the spatio-temporal CSF and luminance adaptation, a novel JND model for videos was proposed in the DCT domain. The experimental results of the subjective quality evaluation tests demonstrate that the proposed JND threshold can hide more noises under the same perceived quality, which is highly consistent with human subjective visual perception.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130316234","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}
In terms of intelligent voice customer service of Inner Mongolia Electric Power, there are a large number of Mongolian speakers. The Mongolian speech recognition in it mainly applies Q&A mode which uses sentences for realizing human-machine dialogue. However, in the process of training the Mongolian acoustic model based on deep neural network-hidden markov model (DNN-HMM), the fragment information of Mongolian speech is mainly applied because of different lengths of speech sentences, it ignores integrity of speech sentences. In this regard, this paper proposes a Mongolian acoustic model based on Bi-directional Long Short-Term Memory-Connectionist Temporal Classification (BLSTM-CTC), which unifies length of input sentences and models complete sentences by inserting BLANK features and labels. The results of comparison experiment of speech recognition between BLSTM-CTC and DNN-HMM shows lower word error rate and sentence error rate of speech recognition based on BLSTM-CTC, especially in later, with reduces by 3.57% and 4.09% respectively. That indicates modeling ability of BLSTM-CTC, especially the modeling ability for sentences, is obviously higher than the DNN-HMM.
{"title":"Research on Mongolian acoustic model based on BLSTM-CTC for Inner Mongolia Electric Power","authors":"Tuya Li, Yaoting Han, Xiaoyu Chen, Sha Li, Yiming Zhao, Shasha Su","doi":"10.1109/cniot55862.2022.00012","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00012","url":null,"abstract":"In terms of intelligent voice customer service of Inner Mongolia Electric Power, there are a large number of Mongolian speakers. The Mongolian speech recognition in it mainly applies Q&A mode which uses sentences for realizing human-machine dialogue. However, in the process of training the Mongolian acoustic model based on deep neural network-hidden markov model (DNN-HMM), the fragment information of Mongolian speech is mainly applied because of different lengths of speech sentences, it ignores integrity of speech sentences. In this regard, this paper proposes a Mongolian acoustic model based on Bi-directional Long Short-Term Memory-Connectionist Temporal Classification (BLSTM-CTC), which unifies length of input sentences and models complete sentences by inserting BLANK features and labels. The results of comparison experiment of speech recognition between BLSTM-CTC and DNN-HMM shows lower word error rate and sentence error rate of speech recognition based on BLSTM-CTC, especially in later, with reduces by 3.57% and 4.09% respectively. That indicates modeling ability of BLSTM-CTC, especially the modeling ability for sentences, is obviously higher than the DNN-HMM.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130898015","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}
In order to improve the accuracy of the monitoring data of marine monitoring sensors, diagnose various potential faults of the sensors and repair them in time, an intelligent data analysis system for diagnosing the faults of the marine monitoring sensor is proposed and developed. Three kinds of data analysis methods based on principle of statistics, including the Grubbs Criterion, the PauTa Criterion and the Dixon Criterion, are used to realize the automatic detection of abnormal data and the corresponding algorithm workflow is designed and implemented with Python intelligent computing modules. Taking the wave sensor data as an example, a set of experiments are conducted to verify the effectiveness of the intelligent system. The results indicate that the system ensures the effectiveness and accuracy of monitoring data properly and it can be used to monitor and analyze the abnormal data of wave sensors for fault diagnosis.
{"title":"An intelligent data analysis system for fault diagnosis of marine monitoring sensors","authors":"Shiyao Zhao, Miaomiao Song, Yunlu Liu, Shuting Liu, Jianjia Zheng, Sijia Wang, Peng Tai, Ziliang Jiang","doi":"10.1109/cniot55862.2022.00044","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00044","url":null,"abstract":"In order to improve the accuracy of the monitoring data of marine monitoring sensors, diagnose various potential faults of the sensors and repair them in time, an intelligent data analysis system for diagnosing the faults of the marine monitoring sensor is proposed and developed. Three kinds of data analysis methods based on principle of statistics, including the Grubbs Criterion, the PauTa Criterion and the Dixon Criterion, are used to realize the automatic detection of abnormal data and the corresponding algorithm workflow is designed and implemented with Python intelligent computing modules. Taking the wave sensor data as an example, a set of experiments are conducted to verify the effectiveness of the intelligent system. The results indicate that the system ensures the effectiveness and accuracy of monitoring data properly and it can be used to monitor and analyze the abnormal data of wave sensors for fault diagnosis.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130995866","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00013
Hezhe Wang
Mobile crowd sensing is a new type of sensing mode and has been widely used. However, mobile crowd sensing requires a large number of users to participate in the execution of sensing tasks, and in the process of executing sensing tasks, users may face the risk of privacy information leakage. Therefore, how to protect the privacy of users is one of the key issues to be solved in the field of mobile crowd sensing. Accordingly, this paper studies the privacy-preserving mechanisms in the field of mobile crowd sensing, and summarizes the existing privacy-preserving research, so as to provide a reference for the in-depth research of mobile crowd sensing.
{"title":"Research on key technologies of mobile crowd sensing for privacy-preserving","authors":"Hezhe Wang","doi":"10.1109/cniot55862.2022.00013","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00013","url":null,"abstract":"Mobile crowd sensing is a new type of sensing mode and has been widely used. However, mobile crowd sensing requires a large number of users to participate in the execution of sensing tasks, and in the process of executing sensing tasks, users may face the risk of privacy information leakage. Therefore, how to protect the privacy of users is one of the key issues to be solved in the field of mobile crowd sensing. Accordingly, this paper studies the privacy-preserving mechanisms in the field of mobile crowd sensing, and summarizes the existing privacy-preserving research, so as to provide a reference for the in-depth research of mobile crowd sensing.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131010560","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00029
Xianpei Luo, Ronghao Yang, Weijing Yao
An accurate and cheap solution to measure distance in 3D space is of great importance in a range of applications. In this paper, we introduce an embedded range measurement system that consists of a core STM89C52 single chip computer, an ultrasonic module, an input module, an alarming module, a display module, an ultrasonic module, and a power module. The corresponding software is also developed with C programming based on these hardware devices. Experiments and tests prove that the proposed system can accurately report the distance in the range between 0.15 meters and 3.7 meters, and the error is less than 0.8% for smooth surfaces. In addition, the proposed device is also cheap and portable and can be used in a range of practical scenarios.
{"title":"Distance Measuring System Based on STM89C52 Single Chip Computer and Ultrasound Sensor","authors":"Xianpei Luo, Ronghao Yang, Weijing Yao","doi":"10.1109/cniot55862.2022.00029","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00029","url":null,"abstract":"An accurate and cheap solution to measure distance in 3D space is of great importance in a range of applications. In this paper, we introduce an embedded range measurement system that consists of a core STM89C52 single chip computer, an ultrasonic module, an input module, an alarming module, a display module, an ultrasonic module, and a power module. The corresponding software is also developed with C programming based on these hardware devices. Experiments and tests prove that the proposed system can accurately report the distance in the range between 0.15 meters and 3.7 meters, and the error is less than 0.8% for smooth surfaces. In addition, the proposed device is also cheap and portable and can be used in a range of practical scenarios.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117088443","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00035
Yueren Sun
With the application of information management system, enterprises build many independent databases to meet different business needs. In order to mine information from multi-source data, data needs to be extracted to a specified data warehouse. This can be accomplished through ETL(extract-transformation-loading), Among them, data extraction is a key stage in ETL. After comparing the advantages and disadvantages of various incremental data extraction mechanisms, this paper proposes a time-stamp-based incremental synchronization scheme, which can not only improve the extraction efficiency, but also solve the shortcoming that the time-stamp extraction method cannot capture deleted data.
{"title":"Research and implementation of an efficient incremental synchronization method based on Timestamp","authors":"Yueren Sun","doi":"10.1109/cniot55862.2022.00035","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00035","url":null,"abstract":"With the application of information management system, enterprises build many independent databases to meet different business needs. In order to mine information from multi-source data, data needs to be extracted to a specified data warehouse. This can be accomplished through ETL(extract-transformation-loading), Among them, data extraction is a key stage in ETL. After comparing the advantages and disadvantages of various incremental data extraction mechanisms, this paper proposes a time-stamp-based incremental synchronization scheme, which can not only improve the extraction efficiency, but also solve the shortcoming that the time-stamp extraction method cannot capture deleted data.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126299197","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}
In this paper, a four-degree-of-freedom flexible endoscope based on tendon drive is proposed, which overcomes the disadvantage of insufficient flexibility at the end of the traditional endoscope, and provides a wider surgical visual field at the same time. This paper focuses on the study of its driving principle and control method. The designed flexible endoscope consists of a flexible arm and a drive box. The flexible segment is composed of 12 skeleton units in each series. The drive box, whose compact body makes the control process more flexible, controls the tightness of the wire through the bidirectional scrolls. Using the model with constant curvature, the motion principle of the flexible endoscope was studied, the mapping relationship among the drive space, the virtual joint space and the workspace was calculated. The drive control system design and drive algorithm programming of the flexible endoscope were completed, and the circular motion and linear motion of the end effector was realized. After experimental verification, the error of the master-slave control accuracy of the end of the four-degree-of-freedom flexible endoscope is less than 3 mm, the workspace is more than 100% larger than that of the traditional endoscope, and the maximum bending angle can reach 140°, realizing the advantage of a larger surgical visual field.
{"title":"Design and application analysis of a novel tendon-drive flexible endoscope","authors":"Qiqi Pan, Hongbo Wang, Jingjing Luo, Fuhao Wang, Wenke Duan, Jiawei Wang, Kangkang Fei, Xiuhong Tang","doi":"10.1109/cniot55862.2022.00038","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00038","url":null,"abstract":"In this paper, a four-degree-of-freedom flexible endoscope based on tendon drive is proposed, which overcomes the disadvantage of insufficient flexibility at the end of the traditional endoscope, and provides a wider surgical visual field at the same time. This paper focuses on the study of its driving principle and control method. The designed flexible endoscope consists of a flexible arm and a drive box. The flexible segment is composed of 12 skeleton units in each series. The drive box, whose compact body makes the control process more flexible, controls the tightness of the wire through the bidirectional scrolls. Using the model with constant curvature, the motion principle of the flexible endoscope was studied, the mapping relationship among the drive space, the virtual joint space and the workspace was calculated. The drive control system design and drive algorithm programming of the flexible endoscope were completed, and the circular motion and linear motion of the end effector was realized. After experimental verification, the error of the master-slave control accuracy of the end of the four-degree-of-freedom flexible endoscope is less than 3 mm, the workspace is more than 100% larger than that of the traditional endoscope, and the maximum bending angle can reach 140°, realizing the advantage of a larger surgical visual field.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134048895","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 : 2022-05-01DOI: 10.1109/cniot55862.2022.00015
Xiaolei Shen, Z. Ge, Quanqin Gao, Haiyang Sun, Xiaoan Tang, Q. Cai
In the virtual-real fusion maintainability test scenario, the point cloud registration of maintenance equipment should meet the online real-time requirements and accuracy requirements, so as to provide a good foundation for virtual-real fusion of equipment. To this end, this paper deeply studies the point cloud registration algorithm based on virtual and real fusion. Firstly, the RGBD depth camera is used to collect depth images and color images of objects (spatial and temporal filtering to remove noise), convert it into point cloud data through the camera’s internal parameters, sampling the CAD (computer aided design) model of the test prototype as a model point cloud. Combined with the internal shape descriptor algorithm (ISS) to collect the feature points of the physical point cloud and the CAD model point cloud, using OpenMP multi-core and multi-thread parallel to accelerate the calculation of normal and calculate the unique shape context (USC) descriptor; Then, sample consensus initial aligment algorithm (SAC-IA) is used to coarsely register the virtual and real point clouds to obtain the initial pose; Finally, iterative nearest point (ICP) algorithm based on point-to-surface is used for precise registration. The experimental results prove that the algorithm of this paper has high efficiency, the point cloud coincidence degree after registration is high, and the error is reduced by nearly one order of magnitude compared with other algorithms, which can provide better algorithm support for virtual and real maintenance field.
{"title":"A point cloud registration algorithm for the fusion of virtual and real maintainability test prototypes","authors":"Xiaolei Shen, Z. Ge, Quanqin Gao, Haiyang Sun, Xiaoan Tang, Q. Cai","doi":"10.1109/cniot55862.2022.00015","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00015","url":null,"abstract":"In the virtual-real fusion maintainability test scenario, the point cloud registration of maintenance equipment should meet the online real-time requirements and accuracy requirements, so as to provide a good foundation for virtual-real fusion of equipment. To this end, this paper deeply studies the point cloud registration algorithm based on virtual and real fusion. Firstly, the RGBD depth camera is used to collect depth images and color images of objects (spatial and temporal filtering to remove noise), convert it into point cloud data through the camera’s internal parameters, sampling the CAD (computer aided design) model of the test prototype as a model point cloud. Combined with the internal shape descriptor algorithm (ISS) to collect the feature points of the physical point cloud and the CAD model point cloud, using OpenMP multi-core and multi-thread parallel to accelerate the calculation of normal and calculate the unique shape context (USC) descriptor; Then, sample consensus initial aligment algorithm (SAC-IA) is used to coarsely register the virtual and real point clouds to obtain the initial pose; Finally, iterative nearest point (ICP) algorithm based on point-to-surface is used for precise registration. The experimental results prove that the algorithm of this paper has high efficiency, the point cloud coincidence degree after registration is high, and the error is reduced by nearly one order of magnitude compared with other algorithms, which can provide better algorithm support for virtual and real maintenance field.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"115 9","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120825560","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}
With the development of intelligent transportation system, license plate recognition technology plays an increasingly important role in our life, and many related technologies have been proposed. However, most of these algorithms have a low success rate in detecting large Angle vehicle images in complex background. In this paper, we propose a Chinese license plate recognition algorithm based on UNet3+ to solve the problem of low license plate recognition rate in large-angle scenarios. The algorithm detects license plate position from three scales to improve the accuracy of license plate position recognition. At the same time, we propose a new loss function to enhance the boundary information of license plate. Finally, on the basis of the Chinese City Parking Dataset (CCPD) Dataset, we added the large Angle vehicle pictures we collected to construct a new Chinese license plate Dataset containing 30,000 images, and increased the amount of data through various data enhancement technologies. Experimental results show that our algorithm is effective in large Angle license plate recognition in complex scenes, and the accuracy rate reaches 94.3%.
{"title":"Chinese License Plate Recognition Algorithm Based On UNet3+","authors":"Menghu Li, Fei Ren, Zhonglin Zhang, Yong Long, Jinquan Zeng","doi":"10.1109/cniot55862.2022.00018","DOIUrl":"https://doi.org/10.1109/cniot55862.2022.00018","url":null,"abstract":"With the development of intelligent transportation system, license plate recognition technology plays an increasingly important role in our life, and many related technologies have been proposed. However, most of these algorithms have a low success rate in detecting large Angle vehicle images in complex background. In this paper, we propose a Chinese license plate recognition algorithm based on UNet3+ to solve the problem of low license plate recognition rate in large-angle scenarios. The algorithm detects license plate position from three scales to improve the accuracy of license plate position recognition. At the same time, we propose a new loss function to enhance the boundary information of license plate. Finally, on the basis of the Chinese City Parking Dataset (CCPD) Dataset, we added the large Angle vehicle pictures we collected to construct a new Chinese license plate Dataset containing 30,000 images, and increased the amount of data through various data enhancement technologies. Experimental results show that our algorithm is effective in large Angle license plate recognition in complex scenes, and the accuracy rate reaches 94.3%.","PeriodicalId":251734,"journal":{"name":"2022 3rd International Conference on Computing, Networks and Internet of Things (CNIOT)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117283542","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：