Pub Date : 2009-06-03DOI: 10.1109/IVS.2009.5164425
Panraphee Raphiphan, A. Zaslavsky, Passakon Prathombutr, P. Meesad
Real time traffic congestion degree is useful information in assisting decision making of drivers. It can also be a factor for calculating other traffic information. The congestion degree can be usually calculated on the basis of sensors installed along roads. It is possible that the sensory data can be lost due to potentially unreliable communication or faulty sensors, leading to lost of important traffic data. In this paper, we propose both adaptive traffic congestion analysis system architecture as well as a novel traffic congestion estimation algorithm that can compensate missing sensory data. An ability to provide traffic condition of road segments at all time is feasible. Unlike other existing methods, our approach aims not to rely on only traffic data from sensors, but utilize discoverable external context instead. The promising experiment result and analysis are reported in this paper. In addition, the context attribute correlation analysis is also discussed.
{"title":"Overcoming uncertainty of roadside sensors with smart adaptive traffic congestion analysis system","authors":"Panraphee Raphiphan, A. Zaslavsky, Passakon Prathombutr, P. Meesad","doi":"10.1109/IVS.2009.5164425","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164425","url":null,"abstract":"Real time traffic congestion degree is useful information in assisting decision making of drivers. It can also be a factor for calculating other traffic information. The congestion degree can be usually calculated on the basis of sensors installed along roads. It is possible that the sensory data can be lost due to potentially unreliable communication or faulty sensors, leading to lost of important traffic data. In this paper, we propose both adaptive traffic congestion analysis system architecture as well as a novel traffic congestion estimation algorithm that can compensate missing sensory data. An ability to provide traffic condition of road segments at all time is feasible. Unlike other existing methods, our approach aims not to rely on only traffic data from sensors, but utilize discoverable external context instead. The promising experiment result and analysis are reported in this paper. In addition, the context attribute correlation analysis is also discussed.","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115296364","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 : 2009-06-03DOI: 10.1109/IVS.2009.5164243
T. Michalke, R. Kastner, Michael J. Herbert, J. Fritsch, C. Goerick
First vision-based approaches for detecting the drivable road area on unmarked streets were introduced in recent years. Although most of these visual feature-based approaches show sound results in scenarios of limited complexity, they seem to lack the necessary system-inherent flexibility to run in complex cluttered environments under changing lighting conditions. Our proposed architecture relies on four novel approaches that make such systems more generic by autonomously adapting important system parameters to the environment. As the presented results show, the approach allows for robust road detection on unmarked inner-city streets without manual tuning of internal parameters. The described system was implemented in C relying on the Intel Performance Primitives library and proved its real-time capability. It will be a sub-module of an advanced driver assistance architecture, which runs in real-time on a test vehicle.
{"title":"Adaptive multi-cue fusion for robust detection of unmarked inner-city streets","authors":"T. Michalke, R. Kastner, Michael J. Herbert, J. Fritsch, C. Goerick","doi":"10.1109/IVS.2009.5164243","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164243","url":null,"abstract":"First vision-based approaches for detecting the drivable road area on unmarked streets were introduced in recent years. Although most of these visual feature-based approaches show sound results in scenarios of limited complexity, they seem to lack the necessary system-inherent flexibility to run in complex cluttered environments under changing lighting conditions. Our proposed architecture relies on four novel approaches that make such systems more generic by autonomously adapting important system parameters to the environment. As the presented results show, the approach allows for robust road detection on unmarked inner-city streets without manual tuning of internal parameters. The described system was implemented in C relying on the Intel Performance Primitives library and proved its real-time capability. It will be a sub-module of an advanced driver assistance architecture, which runs in real-time on a test vehicle.","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124282412","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 : 2009-06-03DOI: 10.1109/IVS.2009.5164308
S. Hold, C. Nunn, A. Kummert, S. Muller-Schneiders
Intelligent Driver Assistance Systems, such as Lane Departure Warning, extract 3D information of the road geometry from a camera. Therefore, the transformation between the image and the ground plane has to be determined with a very high accuracy. Conventional calibration methods are usually a compromise between the accuracy and a preferably small effort for the calibration set-up. In this paper, we present an efficient and robust method for an accurate estimation of the extrinsic parameters based on minimizing an error function. The idea is to avoid the difficult and time-consuming measurement of marker positions in the 3D world coordinate system which is fixed with respect to the vehicle. A pattern of circles is placed on the ground plane in front of the car. For our approach, it is only necessary to measure the relative distances between the centers of the circles to each other. A nonlinear-optimization algorithm minimizes the squared difference between the distances of the backprojected circles segmented in the images on the ground plane and of the measurement in the real world.
{"title":"Efficient and robust extrinsic camera calibration procedure for Lane Departure Warning","authors":"S. Hold, C. Nunn, A. Kummert, S. Muller-Schneiders","doi":"10.1109/IVS.2009.5164308","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164308","url":null,"abstract":"Intelligent Driver Assistance Systems, such as Lane Departure Warning, extract 3D information of the road geometry from a camera. Therefore, the transformation between the image and the ground plane has to be determined with a very high accuracy. Conventional calibration methods are usually a compromise between the accuracy and a preferably small effort for the calibration set-up. In this paper, we present an efficient and robust method for an accurate estimation of the extrinsic parameters based on minimizing an error function. The idea is to avoid the difficult and time-consuming measurement of marker positions in the 3D world coordinate system which is fixed with respect to the vehicle. A pattern of circles is placed on the ground plane in front of the car. For our approach, it is only necessary to measure the relative distances between the centers of the circles to each other. A nonlinear-optimization algorithm minimizes the squared difference between the distances of the backprojected circles segmented in the images on the ground plane and of the measurement in the real world.","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127101949","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 : 2009-06-03DOI: 10.1109/IVS.2009.5164437
W. Wibisono, A. Zaslavsky, Sea Ling
The proliferation of intelligent vehicle, sensor and communication technology has led to the emergence of vehicle-to-vehicle (V2V) applications that aim to increase safety and efficiency of driving. The important requirements of these applications are the capabilities to provide unobtrusive support to the driver and the application capability to adapt to changing situations in the environment without having explicit driver intervention. This paper proposes a new approach for context and situation reasoning in V2V environment. We model context and situations based on Context Spaces and integrate the model with Dempster-Shafer rule of combination for situation reasoning. We also incorporate reliability of each information source in the fusion mechanism based on discount rule. We apply this approach to a context middleware framework that aims to facilitate context and situation reasoning to provide reliable support for cooperative applications in V2V environment and discuss the implementation and experimentation issues of the prototype
{"title":"Improving situation awareness for intelligent on-board vehicle management system using context middleware","authors":"W. Wibisono, A. Zaslavsky, Sea Ling","doi":"10.1109/IVS.2009.5164437","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164437","url":null,"abstract":"The proliferation of intelligent vehicle, sensor and communication technology has led to the emergence of vehicle-to-vehicle (V2V) applications that aim to increase safety and efficiency of driving. The important requirements of these applications are the capabilities to provide unobtrusive support to the driver and the application capability to adapt to changing situations in the environment without having explicit driver intervention. This paper proposes a new approach for context and situation reasoning in V2V environment. We model context and situations based on Context Spaces and integrate the model with Dempster-Shafer rule of combination for situation reasoning. We also incorporate reliability of each information source in the fusion mechanism based on discount rule. We apply this approach to a context middleware framework that aims to facilitate context and situation reasoning to provide reliable support for cooperative applications in V2V environment and discuss the implementation and experimentation issues of the prototype","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"179 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125819559","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 : 2009-06-03DOI: 10.1109/IVS.2009.5164419
Li Xu, Qunfei Zhao, Chengbin Ma, Fangfang Lu
Drivers over the age of 65 are increasing rapidly in numbers and they are inclined to be involved in accidents frequently. In this paper, a multi-source information fusion model to improve the driving safety for older drivers is proposed. First, the influences of surrounding features, such as traffic and weather, on the driving safety are analyzed and the surrounding driving safety degree (SDSD) is proposed to represent it. Second, we analyze the effect of driving behavior characters on the driving safety and name it as the driving behavior safety degree (DBSD). Then we propose a fuzzy information fusion method to evaluate the driver safety degree (DSD) based on the evaluation results of SDSD and DBSD. The fuzzy reasoning rules can be adjusted to satisfy different drivers through analyzing the driving behavior and history traffic accident logs collected. We test our methods based on the data sets collected from the American National Highway Traffic Safety Administration (NHTSA) and the experimental results show that the proposed method is more efficient in improving the older driver's safety.
{"title":"Systematic methodology of multi-source information fusion for improving older drivers' safety","authors":"Li Xu, Qunfei Zhao, Chengbin Ma, Fangfang Lu","doi":"10.1109/IVS.2009.5164419","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164419","url":null,"abstract":"Drivers over the age of 65 are increasing rapidly in numbers and they are inclined to be involved in accidents frequently. In this paper, a multi-source information fusion model to improve the driving safety for older drivers is proposed. First, the influences of surrounding features, such as traffic and weather, on the driving safety are analyzed and the surrounding driving safety degree (SDSD) is proposed to represent it. Second, we analyze the effect of driving behavior characters on the driving safety and name it as the driving behavior safety degree (DBSD). Then we propose a fuzzy information fusion method to evaluate the driver safety degree (DSD) based on the evaluation results of SDSD and DBSD. The fuzzy reasoning rules can be adjusted to satisfy different drivers through analyzing the driving behavior and history traffic accident logs collected. We test our methods based on the data sets collected from the American National Highway Traffic Safety Administration (NHTSA) and the experimental results show that the proposed method is more efficient in improving the older driver's safety.","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"232 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126042569","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 : 2009-06-03DOI: 10.1109/IVS.2009.5164254
Y. Fu, Howard Li, M. Kaye
In this paper, an intelligent vehicle is built using a remote control car. A fuzzy controller is developed for vision based autonomous road following. The analysis and design of fuzzy control laws for steering control of the autonomous nonholonomic robotic vehicle are described. This approach utilizes Lyapunov's direct method to formulate a class of control laws that guarantee the convergence of the steering error. Certain requirements for the control laws are presented for designers to choose a suitable rule base for the fuzzy controller in order to make the system asymptotically stable. Stability of the proposed fuzzy controller is guaranteed theoretically and also demonstrated by simulation studies and experiments. Simulations using the model of the four degree of freedom nonholonomic robotic vehicle are conducted to investigate the performance of the fuzzy controller. The proposed fuzzy controller can achieve the desired steering angle and make the robotic vehicle follow the road successfully. Experiments show that the developed intelligent vehicle is able to follow a mocked road autonomously.
{"title":"Design and stability analysis of a fuzzy controller for autonomous road following","authors":"Y. Fu, Howard Li, M. Kaye","doi":"10.1109/IVS.2009.5164254","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164254","url":null,"abstract":"In this paper, an intelligent vehicle is built using a remote control car. A fuzzy controller is developed for vision based autonomous road following. The analysis and design of fuzzy control laws for steering control of the autonomous nonholonomic robotic vehicle are described. This approach utilizes Lyapunov's direct method to formulate a class of control laws that guarantee the convergence of the steering error. Certain requirements for the control laws are presented for designers to choose a suitable rule base for the fuzzy controller in order to make the system asymptotically stable. Stability of the proposed fuzzy controller is guaranteed theoretically and also demonstrated by simulation studies and experiments. Simulations using the model of the four degree of freedom nonholonomic robotic vehicle are conducted to investigate the performance of the fuzzy controller. The proposed fuzzy controller can achieve the desired steering angle and make the robotic vehicle follow the road successfully. Experiments show that the developed intelligent vehicle is able to follow a mocked road autonomously.","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126941273","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 : 2009-06-03DOI: 10.1109/IVS.2009.5164262
N. Suganuma
The driving support is one of the most important research areas in intelligent transport system (ITS). Moreover, obstacle extraction system is one of most important system. Here, baseline length of our stereovision system is shorter than general one so that the room mirror can cover it. Accordingly, it can be placed unobtrusively and never be unsighted for the driver. Therefore, our stereovision system is suite for interior sensor. In our previous report, we proposed an obstacle extraction method using such stereovision system, and traditional “V-Disparity” approach was extended to more flexible system by using Virtual Disparity Image. Hereby obstacles can be extracted even if the vehicle has large roll movement. In this paper, we will discuss about clustering and tracking method of obstacles.
{"title":"Clustering and tracking of obstacles from Virtual Disparity Image","authors":"N. Suganuma","doi":"10.1109/IVS.2009.5164262","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164262","url":null,"abstract":"The driving support is one of the most important research areas in intelligent transport system (ITS). Moreover, obstacle extraction system is one of most important system. Here, baseline length of our stereovision system is shorter than general one so that the room mirror can cover it. Accordingly, it can be placed unobtrusively and never be unsighted for the driver. Therefore, our stereovision system is suite for interior sensor. In our previous report, we proposed an obstacle extraction method using such stereovision system, and traditional “V-Disparity” approach was extended to more flexible system by using Virtual Disparity Image. Hereby obstacles can be extracted even if the vehicle has large roll movement. In this paper, we will discuss about clustering and tracking method of obstacles.","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"58 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122734355","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 : 2009-06-03DOI: 10.1109/IVS.2009.5164412
Changjiang Zheng, Pan Liu, J. Lu, Hongyun Chen
In this study, a procedure was developed to help transportation professionals evaluate the impacts of U-turns on level of service of signalized intersections using two widely used simulation packages: Synchro and SimTraffic. Two parameters related to U-turns were calibrated based on field data. They are (1) the saturation flow rate for U-turn and left-turn mixed traffic stream; and (2) the turning speeds for U-turning vehicles in the left-turn lane. The saturation flow rate was calculated based on the U-turn adjustment factors developed in a previous study. A linear regression model was developed for predicting the average turning speeds of U-turning vehicles at a signalized intersection given the turning radius provided at the left-turn bay. Delay reported by the calibrated Synchro and SimTraffic models was compared to the average control delay measured in the field. The effectiveness of parameter calibration differs for different signalized intersections. It was generally found that the calibrated Synchro and SimTraffic models provided reasonable delay estimates for U-turns at signalized intersections. With the calibrated and validated Synchro and SimTraffic models, sensitivity analysis was conducted to evaluate how the increased number of U-turning vehicles affects the level of service of signalized intersections. It was found that the sensitivity of the average control delay to changing U-turn volumes differs for different signalized intersections. The sensitivity analysis results suggest that the impacts of U-turns on the level of service of signalized intersections shall be evaluated on a case-by-case basis.
{"title":"Evaluating the effect effects of U-turns on level of service of signalized intersections using synchro and SimTraffic","authors":"Changjiang Zheng, Pan Liu, J. Lu, Hongyun Chen","doi":"10.1109/IVS.2009.5164412","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164412","url":null,"abstract":"In this study, a procedure was developed to help transportation professionals evaluate the impacts of U-turns on level of service of signalized intersections using two widely used simulation packages: Synchro and SimTraffic. Two parameters related to U-turns were calibrated based on field data. They are (1) the saturation flow rate for U-turn and left-turn mixed traffic stream; and (2) the turning speeds for U-turning vehicles in the left-turn lane. The saturation flow rate was calculated based on the U-turn adjustment factors developed in a previous study. A linear regression model was developed for predicting the average turning speeds of U-turning vehicles at a signalized intersection given the turning radius provided at the left-turn bay. Delay reported by the calibrated Synchro and SimTraffic models was compared to the average control delay measured in the field. The effectiveness of parameter calibration differs for different signalized intersections. It was generally found that the calibrated Synchro and SimTraffic models provided reasonable delay estimates for U-turns at signalized intersections. With the calibrated and validated Synchro and SimTraffic models, sensitivity analysis was conducted to evaluate how the increased number of U-turning vehicles affects the level of service of signalized intersections. It was found that the sensitivity of the average control delay to changing U-turn volumes differs for different signalized intersections. The sensitivity analysis results suggest that the impacts of U-turns on the level of service of signalized intersections shall be evaluated on a case-by-case basis.","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129566157","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 : 2009-06-03DOI: 10.1109/IVS.2009.5164426
Jin Li-sheng, Fang Wen-ping, ZHang Ying-nan, Yang Shuang-bin, Hou Hai-jing
To solve those traffic accidents during conscious lane change of vehicles on highway under dangerous conditions, a new safety lane change model is established on the basis of the lane departure warning system which Jilin University has developed. This model is studied under a typical scenario on highway, and also considered with the actual driver behavior that most vehicles always accelerate during lane changing process. Simulation software is developed to testify the model performance based on Matlab7.0. The results show that the new lane change model can offer certain technical foundation for actualization of safety lane change system of vehicle assistant driving on highway.
{"title":"Research on safety lane change model of driver assistant system on highway","authors":"Jin Li-sheng, Fang Wen-ping, ZHang Ying-nan, Yang Shuang-bin, Hou Hai-jing","doi":"10.1109/IVS.2009.5164426","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164426","url":null,"abstract":"To solve those traffic accidents during conscious lane change of vehicles on highway under dangerous conditions, a new safety lane change model is established on the basis of the lane departure warning system which Jilin University has developed. This model is studied under a typical scenario on highway, and also considered with the actual driver behavior that most vehicles always accelerate during lane changing process. Simulation software is developed to testify the model performance based on Matlab7.0. The results show that the new lane change model can offer certain technical foundation for actualization of safety lane change system of vehicle assistant driving on highway.","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130385595","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 : 2009-06-03DOI: 10.1109/IVS.2009.5164345
Matthias Serfling, O. Loehlein, R. Schweiger, K. Dietmayer
This contribution presents a robust pedestrian detection system at night that fuses a camera sensor and a scanning radar sensor on feature level. Each sensor defines an overdetermined set of features to be selected and parameterized using the supervised training algorithm AdaBoost. This technique assures an optimal selection and weighting of the features from both sensors depending on their discriminative power for the classification task. In the radar plane a new complex signal filter has been derived which describes a local similarity measure of velocity differences. In order to achieve realtime capability multiple classifiers are combined using a cascade.
{"title":"Camera and imaging radar feature level sensorfusion for night vision pedestrian recognition","authors":"Matthias Serfling, O. Loehlein, R. Schweiger, K. Dietmayer","doi":"10.1109/IVS.2009.5164345","DOIUrl":"https://doi.org/10.1109/IVS.2009.5164345","url":null,"abstract":"This contribution presents a robust pedestrian detection system at night that fuses a camera sensor and a scanning radar sensor on feature level. Each sensor defines an overdetermined set of features to be selected and parameterized using the supervised training algorithm AdaBoost. This technique assures an optimal selection and weighting of the features from both sensors depending on their discriminative power for the classification task. In the radar plane a new complex signal filter has been derived which describes a local similarity measure of velocity differences. In order to achieve realtime capability multiple classifiers are combined using a cascade.","PeriodicalId":396749,"journal":{"name":"2009 IEEE Intelligent Vehicles Symposium","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123943515","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：