Chong, Weng Kong (2019) Design And Implementation Of Deep Learning Enabled Mobile Road Safety Inspection System. Project Report. UTeM, Melaka, Malaysia. (Submitted)
|
Text
Design and implementation of deep learning enabled mobile road safety inspection system.pdf Download (533kB) | Preview |
Abstract
An AI-enabled mobile road safety inspection system based on Deep Neural Network (DNN) is proposed in this work. A complete mobile road safety inspection system takes into account of many aspects such as road condition, traffic volume, road curvature for road safety inspection and this study focuses solely on the speed of vehicles surrounding the road safety survey vehicle travelling along the road. The developed system consists of Intel OpenVINO Inference Engine for deep neural network vehicle detection, Deep SORT tracker for vehicle tracking, and planar homography based vehicle distance and speed estimation. The vehicle speed detection system is then evaluated using three different cameras in order to determine best optimized field of view and speed measurement accuracy. Results of evaluation show that automatic speed estimation with DNN achieve accuracy with −2.08±5.24
| Item Type: | Final Year Project (Project Report) |
|---|---|
| Uncontrolled Keywords: | Image processing, Optical pattern recognition |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) |
| Divisions: | Library > Final Year Project > FKEKK |
| Depositing User: | Sabariah Ismail |
| Date Deposited: | 26 Jun 2020 02:11 |
| Last Modified: | 18 Aug 2020 06:04 |
| URI: | http://digitalcollection.utem.edu.my/id/eprint/24384 |
Actions (login required)
 |
View Item |
Altmetric
Altmetric