Pub Date : 2021-10-29DOI: 10.1109/ICEMI52946.2021.9679522
Ju Xiao, Junfeng Zhao, Fei Wu, Chengcheng Xie, Tao Chen, Hongliu Liu, Ying Chen
To solve the problems of limited range and low measurement accuracy of traditional reversing radar, a high-precision radar detection system based on dual micro-control units is designed. In this system, two STC89C52 single-chip microcomputers are used to control eight-channel ultrasonic sensors to carry out ranging to avoid blind area. Meanwhile, a temperature sensor is used to detect the environment temperature in real time, and temperature compensation ultrasonic wave velocity is applied to improve the measurement accuracy. In addition, 5 buttons are applied to set alarm threshold of buzzer which can contorl buzzer's frequency varies with different frequencies by different distances. Moreover, the design used LCD1602 and light-emitting diode for display and further warning. The test results show that the linear ranging distance of the system is 0.01 ~ 7 m, and the ranging accuracy can reach 0.01 m.
{"title":"Design of Ultrasonic Radar Detection System","authors":"Ju Xiao, Junfeng Zhao, Fei Wu, Chengcheng Xie, Tao Chen, Hongliu Liu, Ying Chen","doi":"10.1109/ICEMI52946.2021.9679522","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679522","url":null,"abstract":"To solve the problems of limited range and low measurement accuracy of traditional reversing radar, a high-precision radar detection system based on dual micro-control units is designed. In this system, two STC89C52 single-chip microcomputers are used to control eight-channel ultrasonic sensors to carry out ranging to avoid blind area. Meanwhile, a temperature sensor is used to detect the environment temperature in real time, and temperature compensation ultrasonic wave velocity is applied to improve the measurement accuracy. In addition, 5 buttons are applied to set alarm threshold of buzzer which can contorl buzzer's frequency varies with different frequencies by different distances. Moreover, the design used LCD1602 and light-emitting diode for display and further warning. The test results show that the linear ranging distance of the system is 0.01 ~ 7 m, and the ranging accuracy can reach 0.01 m.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132248914","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 : 2021-10-29DOI: 10.1109/ICEMI52946.2021.9679671
Weidi Kong, Haitao Zhang, Wei Zhao, Kai Tang
Canny edge detection based on optimization algorithm has the advantages of high accuracy and high signal-to-noise ratio. However, because the Canny algorithm was based on gray image, some edge information will be lost during the process of converting color image to gray image. Aiming at this problem, the edge detection method of color image based on vector properties was proposed in this paper. According to the properties of vectors with direction and length, the image gradient calculation of Canny algorithm was improved from the angle and length of vector, in order to realize edge feature detection of color image directly. Compared with the results detected by the conventional Canny algorithm, this method can get clearer and more accurate edges. When combined with specific color markers in specific application scenarios, this method can obtain ideal detection results that only contains the edge of the marker.
{"title":"Research on Canny Edge Feature Detection Technology of Color Image Based on Vector Properties","authors":"Weidi Kong, Haitao Zhang, Wei Zhao, Kai Tang","doi":"10.1109/ICEMI52946.2021.9679671","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679671","url":null,"abstract":"Canny edge detection based on optimization algorithm has the advantages of high accuracy and high signal-to-noise ratio. However, because the Canny algorithm was based on gray image, some edge information will be lost during the process of converting color image to gray image. Aiming at this problem, the edge detection method of color image based on vector properties was proposed in this paper. According to the properties of vectors with direction and length, the image gradient calculation of Canny algorithm was improved from the angle and length of vector, in order to realize edge feature detection of color image directly. Compared with the results detected by the conventional Canny algorithm, this method can get clearer and more accurate edges. When combined with specific color markers in specific application scenarios, this method can obtain ideal detection results that only contains the edge of the marker.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132386225","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 : 2021-10-29DOI: 10.1109/ICEMI52946.2021.9679498
Suyang Liu, Jun Yang, Xiye Guo, Kai Liu
Precise clock synchronization is indispensable in many scientific and industrial applications, such as timing laboratories, global satellite navigation systems, and distributed collaborative systems, etc. Digital locked loop (DPLL) is an efficient method to realize real-time and precise clock synchronization. One of the problems when the DPLL is used is to select its coefficients for minimum synchronization errors. This paper proposes an optimized DPLL synchronization method which is equivalent to a steady-state Kalman filter. The coefficients of the DPLL are decided by the gain of the Kalman filter. Simulation shows that the proposed method approaches the minimum synchronization error by comparing to the classical DPLL with different bandwidth. The proposed method has less computation demand, but achieves the same synchronization precision as the Kalman filter.
{"title":"A Clock Synchronization Method Based on An Optimized DPLL Equivalent to A Steady State Kalman Filter","authors":"Suyang Liu, Jun Yang, Xiye Guo, Kai Liu","doi":"10.1109/ICEMI52946.2021.9679498","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679498","url":null,"abstract":"Precise clock synchronization is indispensable in many scientific and industrial applications, such as timing laboratories, global satellite navigation systems, and distributed collaborative systems, etc. Digital locked loop (DPLL) is an efficient method to realize real-time and precise clock synchronization. One of the problems when the DPLL is used is to select its coefficients for minimum synchronization errors. This paper proposes an optimized DPLL synchronization method which is equivalent to a steady-state Kalman filter. The coefficients of the DPLL are decided by the gain of the Kalman filter. Simulation shows that the proposed method approaches the minimum synchronization error by comparing to the classical DPLL with different bandwidth. The proposed method has less computation demand, but achieves the same synchronization precision as the Kalman filter.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133740645","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 : 2021-10-29DOI: 10.1109/ICEMI52946.2021.9679676
Ying Cai
In order to solve the calibration problem of the laser energy meter, a laser energy meter calibration system was designed. The system should be composed of laser source, beam splitter, beam expander, monitoring energy meter, standard energy meter, electric translation stage, optical platform, computer, data acquisition and analysis processing system and optical path installation accessories. The comparison method was used calibration. It is proposed to improve the technical indicators of the system from two aspects: solving the traceability of the laser energy meter calibration system and eliminating the influence of the change of laser energy over time on the measurement results. The measurement uncertainty of laser energy calibration is analyzed in detail. The system realizes a wide range of laser energy ranging from 10mJ to 40J, with a relative extended uncertainty of 2% ($k=2$), and directly traces the laser energy to the National Sub-standard of Pulsed Laser Energy. The system was used to calibrate and verify the E1000 40J laser energy meter of the Chinese Academy of Metrology and the 3SIGMA/J50-110-25ST laser energy meter of Coherent. The test results show that the laser energy meter calibration device can be used for the measurement calibration of the laser energy meter with $1.064mu mathrm{m}$ and the power measurement uncertainty below 2% ($k=2$). The establishment method and testing technology of the system can provide reference for the establishment of laser energy meter measurement standards.
{"title":"Design of Laser Energy Meter Calibration System","authors":"Ying Cai","doi":"10.1109/ICEMI52946.2021.9679676","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679676","url":null,"abstract":"In order to solve the calibration problem of the laser energy meter, a laser energy meter calibration system was designed. The system should be composed of laser source, beam splitter, beam expander, monitoring energy meter, standard energy meter, electric translation stage, optical platform, computer, data acquisition and analysis processing system and optical path installation accessories. The comparison method was used calibration. It is proposed to improve the technical indicators of the system from two aspects: solving the traceability of the laser energy meter calibration system and eliminating the influence of the change of laser energy over time on the measurement results. The measurement uncertainty of laser energy calibration is analyzed in detail. The system realizes a wide range of laser energy ranging from 10mJ to 40J, with a relative extended uncertainty of 2% ($k=2$), and directly traces the laser energy to the National Sub-standard of Pulsed Laser Energy. The system was used to calibrate and verify the E1000 40J laser energy meter of the Chinese Academy of Metrology and the 3SIGMA/J50-110-25ST laser energy meter of Coherent. The test results show that the laser energy meter calibration device can be used for the measurement calibration of the laser energy meter with $1.064mu mathrm{m}$ and the power measurement uncertainty below 2% ($k=2$). The establishment method and testing technology of the system can provide reference for the establishment of laser energy meter measurement standards.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114612821","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 : 2021-10-29DOI: 10.1109/ICEMI52946.2021.9679530
Changchang Zhao, H. Liang, J. Ge
Positron Annihilation Lifetime Spectroscopy (PALS) is a non-destructive spectroscopy technique that can study various phenomena and material properties on an atomic scale. As the traditional analog PALS spectrometers built with multiple nuclear instrument modules gradually showed performance defects due to accuracy limitations and long-term instability, the digital PALS spectrometers began to become popular. However, most digital PALS spectrometers based on oscilloscopes or digitizers are expensive. In order to reduce costs while maintaining high rate counts, this paper proposes an electronic design for a multichannel coincidence digital positron annihilation lifetime spectrometer. Eight identical signal channels are used in the design. Each channel contains an energy measurement circuit and a timing circuit. The analog front-end electronics is designed, simulated, manufactured, and tested. The results show that it has fine linear amplification, well shaping function, low output noise, and sufficient bandwidth. After the firmware and software development is completed, further testing and performance verification will be carried out.
{"title":"Electronic Design for Multichannel Coincidence Digital Positron Annihilation Lifetime Spectrometer","authors":"Changchang Zhao, H. Liang, J. Ge","doi":"10.1109/ICEMI52946.2021.9679530","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679530","url":null,"abstract":"Positron Annihilation Lifetime Spectroscopy (PALS) is a non-destructive spectroscopy technique that can study various phenomena and material properties on an atomic scale. As the traditional analog PALS spectrometers built with multiple nuclear instrument modules gradually showed performance defects due to accuracy limitations and long-term instability, the digital PALS spectrometers began to become popular. However, most digital PALS spectrometers based on oscilloscopes or digitizers are expensive. In order to reduce costs while maintaining high rate counts, this paper proposes an electronic design for a multichannel coincidence digital positron annihilation lifetime spectrometer. Eight identical signal channels are used in the design. Each channel contains an energy measurement circuit and a timing circuit. The analog front-end electronics is designed, simulated, manufactured, and tested. The results show that it has fine linear amplification, well shaping function, low output noise, and sufficient bandwidth. After the firmware and software development is completed, further testing and performance verification will be carried out.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115071933","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 : 2021-10-29DOI: 10.1109/ICEMI52946.2021.9679526
Zhiyu Qu, Shuwang Cai, Qingbo Ji, Lingjing Xu
The size of urine sediment image is small, different categories are easy to be confused, and feature extraction is difficult. This paper proposes an automatic recognition method of urine sediment images based on hourglass residual structure and super-resolution image reconstruction. First, annotate and preprocess the urine sediment image to generate a urine sediment data set. Then, the super-resolution reconstruction technology is used to reconstruct the small-size urine sediment image to adapt to the input of the deep learning model. Finally, an hourglass residual network is constructed to automatically extract the features of the urine sediment image to realize the classification and recognition of the urine sediment image. The experimental results show that the overall accuracy of the method for the recognition of 13 kinds of urine sediment images can reach 99.05%. This method is lightweight enough while maintaining the depth of the network. The number of parameters is 0.73M, which is conducive to porting to mobile devices. This paper proposes a new intelligent recognition method for urine sediment images, which has a good prospect for engineering applications.
{"title":"Lightweight Urine Sediment Image Recognition Network Based on Deep Separable Residual Structure","authors":"Zhiyu Qu, Shuwang Cai, Qingbo Ji, Lingjing Xu","doi":"10.1109/ICEMI52946.2021.9679526","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679526","url":null,"abstract":"The size of urine sediment image is small, different categories are easy to be confused, and feature extraction is difficult. This paper proposes an automatic recognition method of urine sediment images based on hourglass residual structure and super-resolution image reconstruction. First, annotate and preprocess the urine sediment image to generate a urine sediment data set. Then, the super-resolution reconstruction technology is used to reconstruct the small-size urine sediment image to adapt to the input of the deep learning model. Finally, an hourglass residual network is constructed to automatically extract the features of the urine sediment image to realize the classification and recognition of the urine sediment image. The experimental results show that the overall accuracy of the method for the recognition of 13 kinds of urine sediment images can reach 99.05%. This method is lightweight enough while maintaining the depth of the network. The number of parameters is 0.73M, which is conducive to porting to mobile devices. This paper proposes a new intelligent recognition method for urine sediment images, which has a good prospect for engineering applications.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131855794","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 : 2021-10-29DOI: 10.1109/ICEMI52946.2021.9679525
Xingen Liu, Xia Li, Lingshan Yang, Weihui Zhou, Yang Xiang
In this paper, the optimal installation position of pitot tube for measuring L-type pitot coefficient in wind tunnel is studied. By analyzing the calculation formula of pitot tube coefficient, it is concluded that the influencing factor of pitot tube coefficient measurement is the difference between pitot tube total pressure and static pressure. The CFD simulation of the wind speed field near the pitot tube in the wind tunnel shows that the position that may affect the measurement results is close to the side wall of the wind tunnel. Experiments are carried out at these positions. The pitot tube is installed at different positions in the wind tunnel, the differential pressure value is measured, and the area with small differential pressure change is observed. It is concluded that the optimal installation position for pitot tube measurement is the middle position more than 50mm away from the standard pitot tube and more than 50mm away from the side wall of the wind tunnel.
{"title":"Study on the Optimal Installation Position of L-type Pitot Tube During Calibration","authors":"Xingen Liu, Xia Li, Lingshan Yang, Weihui Zhou, Yang Xiang","doi":"10.1109/ICEMI52946.2021.9679525","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679525","url":null,"abstract":"In this paper, the optimal installation position of pitot tube for measuring L-type pitot coefficient in wind tunnel is studied. By analyzing the calculation formula of pitot tube coefficient, it is concluded that the influencing factor of pitot tube coefficient measurement is the difference between pitot tube total pressure and static pressure. The CFD simulation of the wind speed field near the pitot tube in the wind tunnel shows that the position that may affect the measurement results is close to the side wall of the wind tunnel. Experiments are carried out at these positions. The pitot tube is installed at different positions in the wind tunnel, the differential pressure value is measured, and the area with small differential pressure change is observed. It is concluded that the optimal installation position for pitot tube measurement is the middle position more than 50mm away from the standard pitot tube and more than 50mm away from the side wall of the wind tunnel.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"254 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132182969","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 : 2021-10-29DOI: 10.1109/ICEMI52946.2021.9679634
Jun He, Juan Chen
There are many problems with traditional solid-state ion electrodes, such as unstable, long response time, etc. In this paper, a glassy carbon electrode was used as the conductor, and poly-ethylenedioxythiophene (PEDOT) was used as the conductive polymer layer. A new type of trace ion all-solid-state electrode was prepared by combining the sensitive membrane of the working electrode with a drop-coating method. In order to meet the needs of soft water online detection, which is important to the marine nuclear power plant, chloride ion working electrode and nitrate ion working electrode had been developed. The electrode capacitance, resistance characteristics, repeatability, detection limit, response time, selectivity coefficient, and temperature influence have been experimented. Compared with the solid station electrode without conductive polymer layer, the electrode with conductive polymer improved resistance and capacitance characteristics, improved stability and repeatability, reduced response time. The average detection error of ion select electrodes on soft water was less than 5% according to the result.
{"title":"Research on the Response Characteristics of Conductive Polymerization of Selective Electrode for Marine Pure Water","authors":"Jun He, Juan Chen","doi":"10.1109/ICEMI52946.2021.9679634","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679634","url":null,"abstract":"There are many problems with traditional solid-state ion electrodes, such as unstable, long response time, etc. In this paper, a glassy carbon electrode was used as the conductor, and poly-ethylenedioxythiophene (PEDOT) was used as the conductive polymer layer. A new type of trace ion all-solid-state electrode was prepared by combining the sensitive membrane of the working electrode with a drop-coating method. In order to meet the needs of soft water online detection, which is important to the marine nuclear power plant, chloride ion working electrode and nitrate ion working electrode had been developed. The electrode capacitance, resistance characteristics, repeatability, detection limit, response time, selectivity coefficient, and temperature influence have been experimented. Compared with the solid station electrode without conductive polymer layer, the electrode with conductive polymer improved resistance and capacitance characteristics, improved stability and repeatability, reduced response time. The average detection error of ion select electrodes on soft water was less than 5% according to the result.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133948863","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 : 2021-10-29DOI: 10.1109/ICEMI52946.2021.9679539
Jiangmiao Zhu, Xie Dong, Pengfei Wang, Yan Huang
In view of the problem of inaccurate navigation pattern recognition in complex driving environment, a navigation pattern recognition algorithm based on random forest is proposed. Firstly, the error source of navigation components is analyzed to determine the factors affecting the accuracy of different navigation patterns, as the characteristic vector of designing random forest models. Secondly, the data set is constructed by recording the data collected by the vehicle. The random forest model is trained with 70% random data in data set, and the recognition rate is verified with the remaining 30% data. The average recognition rate of this algorithm is 88%, which can meet the needs of the work.
{"title":"An Algorithm Based on Random Forest for In-vehicle Navigation Pattern Recognition","authors":"Jiangmiao Zhu, Xie Dong, Pengfei Wang, Yan Huang","doi":"10.1109/ICEMI52946.2021.9679539","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679539","url":null,"abstract":"In view of the problem of inaccurate navigation pattern recognition in complex driving environment, a navigation pattern recognition algorithm based on random forest is proposed. Firstly, the error source of navigation components is analyzed to determine the factors affecting the accuracy of different navigation patterns, as the characteristic vector of designing random forest models. Secondly, the data set is constructed by recording the data collected by the vehicle. The random forest model is trained with 70% random data in data set, and the recognition rate is verified with the remaining 30% data. The average recognition rate of this algorithm is 88%, which can meet the needs of the work.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132594678","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}
Aiming at the problem of incomplete segmentation of target contour information in traditional pedestrian target segmentation methods under different gaits such as walking and jogging, this paper presents a segmentation method for pedestrian objects in ground point cloud image. Firstly, statistical outlier removal combined with the random sample consensus method is proposed to remove outliers and segment all plane point clouds. Then, based on the ground point cloud image, a range maximum search method is proposed to extract the region of interest. Finally, the pass-through filter is used to segment pedestrian targets in the region of interest. In this paper, point cloud data of four gaits of three pedestrians are collected and compared with traditional Euclidean Clustering and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) Clustering. The results show that the average accuracy of pedestrian target segmentation under different gait is 92.96%. The validity and advance of the proposed algorithm are proved.
{"title":"Segmentation Method of Pedestrian Object in Ground Point Cloud Image","authors":"Xin Shi, Liang Yu, Pengjie Qin, Zhirui Fan, Fei Liang, Gaojie He","doi":"10.1109/ICEMI52946.2021.9679548","DOIUrl":"https://doi.org/10.1109/ICEMI52946.2021.9679548","url":null,"abstract":"Aiming at the problem of incomplete segmentation of target contour information in traditional pedestrian target segmentation methods under different gaits such as walking and jogging, this paper presents a segmentation method for pedestrian objects in ground point cloud image. Firstly, statistical outlier removal combined with the random sample consensus method is proposed to remove outliers and segment all plane point clouds. Then, based on the ground point cloud image, a range maximum search method is proposed to extract the region of interest. Finally, the pass-through filter is used to segment pedestrian targets in the region of interest. In this paper, point cloud data of four gaits of three pedestrians are collected and compared with traditional Euclidean Clustering and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) Clustering. The results show that the average accuracy of pedestrian target segmentation under different gait is 92.96%. The validity and advance of the proposed algorithm are proved.","PeriodicalId":289132,"journal":{"name":"2021 IEEE 15th International Conference on Electronic Measurement & Instruments (ICEMI)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129359783","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：