Road User Position and Speed Estimation via Deep Learning from Calibrated Fisheye Videos
In this paper, we present a deep learning processing flow aimed at Advanced Driving Assistance Systems (ADASs) for urban road users. We use a fine analysis of the optical setup of a fisheye camera and present a detailed procedure to obtain Global Navigation Satellite System (GNSS) coordinates along...
| Hlavní autoři: | , , , , |
|---|---|
| Médium: | Článek |
| Jazyk: | English |
| Vydáno: |
MDPI AG
2023-02-01
|
| Edice: | Sensors |
| Témata: | |
| On-line přístup: | https://www.mdpi.com/1424-8220/23/5/2637 |
| _version_ | 1827752119740923904 |
|---|---|
| author | Yves Berviller Masoomeh Shireen Ansarnia Etienne Tisserand Patrick Schweitzer Alain Tremeau |
| author_facet | Yves Berviller Masoomeh Shireen Ansarnia Etienne Tisserand Patrick Schweitzer Alain Tremeau |
| author_sort | Yves Berviller |
| collection | DOAJ |
| description | In this paper, we present a deep learning processing flow aimed at Advanced Driving Assistance Systems (ADASs) for urban road users. We use a fine analysis of the optical setup of a fisheye camera and present a detailed procedure to obtain Global Navigation Satellite System (GNSS) coordinates along with the speed of the moving objects. The camera to world transform incorporates the lens distortion function. YOLOv4, re-trained with ortho-photographic fisheye images, provides road user detection. All the information extracted from the image by our system represents a small payload and can easily be broadcast to the road users. The results show that our system is able to properly classify and localize the detected objects in real time, even in low-light-illumination conditions. For an effective observation area of 20 m × 50 m, the error of the localization is in the order of one meter. Although an estimation of the velocities of the detected objects is carried out by offline processing with the FlowNet2 algorithm, the accuracy is quite good, with an error below one meter per second for urban speed range (0 to 15 m/s). Moreover, the almost ortho-photographic configuration of the imaging system ensures that the anonymity of all street users is guaranteed. |
| first_indexed | 2024-03-11T07:10:17Z |
| format | Article |
| id | doaj.art-4abaac2efe0842e1818d832b51a6252b |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T07:10:17Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-4abaac2efe0842e1818d832b51a6252b2023-11-17T08:37:33ZengMDPI AGSensors1424-82202023-02-01235263710.3390/s23052637Road User Position and Speed Estimation via Deep Learning from Calibrated Fisheye VideosYves Berviller0Masoomeh Shireen Ansarnia1Etienne Tisserand2Patrick Schweitzer3Alain Tremeau4Institut Jean Lamour, Université de Lorraine, UMR7198, F-54052 Nancy, FranceInstitut Jean Lamour, Université de Lorraine, UMR7198, F-54052 Nancy, FranceInstitut Jean Lamour, Université de Lorraine, UMR7198, F-54052 Nancy, FranceInstitut Jean Lamour, Université de Lorraine, UMR7198, F-54052 Nancy, FranceIndependent Researcher, F-57155 Marly, FranceIn this paper, we present a deep learning processing flow aimed at Advanced Driving Assistance Systems (ADASs) for urban road users. We use a fine analysis of the optical setup of a fisheye camera and present a detailed procedure to obtain Global Navigation Satellite System (GNSS) coordinates along with the speed of the moving objects. The camera to world transform incorporates the lens distortion function. YOLOv4, re-trained with ortho-photographic fisheye images, provides road user detection. All the information extracted from the image by our system represents a small payload and can easily be broadcast to the road users. The results show that our system is able to properly classify and localize the detected objects in real time, even in low-light-illumination conditions. For an effective observation area of 20 m × 50 m, the error of the localization is in the order of one meter. Although an estimation of the velocities of the detected objects is carried out by offline processing with the FlowNet2 algorithm, the accuracy is quite good, with an error below one meter per second for urban speed range (0 to 15 m/s). Moreover, the almost ortho-photographic configuration of the imaging system ensures that the anonymity of all street users is guaranteed.https://www.mdpi.com/1424-8220/23/5/2637ADASI2Vdeep learningcamera to world transformreal time |
| spellingShingle | Yves Berviller Masoomeh Shireen Ansarnia Etienne Tisserand Patrick Schweitzer Alain Tremeau Road User Position and Speed Estimation via Deep Learning from Calibrated Fisheye Videos Sensors ADAS I2V deep learning camera to world transform real time |
| title | Road User Position and Speed Estimation via Deep Learning from Calibrated Fisheye Videos |
| title_full | Road User Position and Speed Estimation via Deep Learning from Calibrated Fisheye Videos |
| title_fullStr | Road User Position and Speed Estimation via Deep Learning from Calibrated Fisheye Videos |
| title_full_unstemmed | Road User Position and Speed Estimation via Deep Learning from Calibrated Fisheye Videos |
| title_short | Road User Position and Speed Estimation via Deep Learning from Calibrated Fisheye Videos |
| title_sort | road user position and speed estimation via deep learning from calibrated fisheye videos |
| topic | ADAS I2V deep learning camera to world transform real time |
| url | https://www.mdpi.com/1424-8220/23/5/2637 |
| work_keys_str_mv | AT yvesberviller roaduserpositionandspeedestimationviadeeplearningfromcalibratedfisheyevideos AT masoomehshireenansarnia roaduserpositionandspeedestimationviadeeplearningfromcalibratedfisheyevideos AT etiennetisserand roaduserpositionandspeedestimationviadeeplearningfromcalibratedfisheyevideos AT patrickschweitzer roaduserpositionandspeedestimationviadeeplearningfromcalibratedfisheyevideos AT alaintremeau roaduserpositionandspeedestimationviadeeplearningfromcalibratedfisheyevideos |