Region of Interest Prediction Method under Dynamic Characteristics of Vehicles

This paper proposes a region of interest (ROI) prediction method that considers the dynamic characteristics of vehicles to address the inaccuracy of ROI detection with the lateral movement of vehicles. The scene and traffic system models are established through PreScan. An ROI prediction algorithm i...

Full description

Saved in:
Bibliographic Details
Published in:Chinese Automation Congress (Online) pp. 4268 - 4273
Main Authors: Li, Shaosong, Tian, Yunsheng, Li, Zheng, Yu, Zhixin, Zhang, Bangcheng, Cui, Gaojian
Format: Conference Proceeding
Language:English
Published: IEEE 01.11.2019
Subjects:
Aerodynamics
coordinate transformation
Kalman filter
Kalman filters
Mathematical model
Predictive models
PreScan
ROI prediction
Vehicle dynamics
vehicle status
ISSN:2688-0938
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper proposes a region of interest (ROI) prediction method that considers the dynamic characteristics of vehicles to address the inaccuracy of ROI detection with the lateral movement of vehicles. The scene and traffic system models are established through PreScan. An ROI prediction algorithm is designed in geodetic coordinate systems based on the system model and Kalman filter theory. The ROI is then converted into the pixel coordinate system through coordinate transformation. The proposed method can detect lane lines in the area in real-time. Simulation experiments under different working conditions show that the proposed method can accurately predict the ROI region and adjust the ROI position in real-time according to the vehicle motion state.
ISSN:2688-0938
DOI:10.1109/CAC48633.2019.8996194