Please use this identifier to cite or link to this item: http://hdl.handle.net/2080/3359
Full metadata record
DC FieldValueLanguage
dc.contributor.authorGauttam, Hutesh Kumar-
dc.contributor.authorMohapatra, Ramesh Kumar-
dc.date.accessioned2019-10-04T06:24:17Z-
dc.date.available2020-10-04T06:24:17Z-
dc.date.issued2019-09-
dc.identifier.citation4th International conference on computer vision & image processing (CVIP 2019),Jaipur, India, 27-29 September 2019en_US
dc.identifier.urihttp://hdl.handle.net/2080/3359-
dc.descriptionCopyright of this document belongs to proceedings publisher.en_US
dc.description.abstract. Vehicle accidents are increasing day by day as a result of high-speed vehicles on highways so, the speed determination of fastapproaching vehicles is becoming a challenging task with moving camera. Most of the vehicles are driven above the prescribed vehicle speed. On expressways, light motor vehicles are unaware of the speed of the rapid vehicle following to them. So in this paper, an algorithm has been proposed to anticipate the speed of the fast approaching vehicle by a moving camera to offer better security. The proposed method comprises of mainly hree successive steps, vehicle detection using YOLO (You Only Look Once) algorithm on the video stream, vehicle position tracking over the continuous frame and speed calculation of approaching vehicle using a moving camera. The relative speed is determined using relative distance travelled by vehicle over a number of frames. This proposed algorithm is giving on an average 90% accuracy in speed prediction of approaching vehicles.en_US
dc.subjectSpeed Detection Cameraen_US
dc.subjectVehicle Detectionen_US
dc.subjectVehicle Trackingen_US
dc.subjectMachine Learning Techniques (YOLO Algorithm)en_US
dc.titleSpeed Prediction of Fast Approaching Vehicle Using Moving Cameraen_US
dc.typeArticleen_US
Appears in Collections:Conference Papers

Files in This Item:
File Description SizeFormat 
2019_CVIP_RKMohapatra_SpeedPrediction.pdfConference paper1.08 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.