Research And Implementation Of Forward Collision Warning System Based On IOS Platform

With the continuous improvement of people's living standards,the number of cars on the road has been increasing,while traffic accidents occur frequently.In recent years,traffic accidents have become a serious social problem.In order to travel safely and reduce the occurrence of traffic accidents,vehicle safety technology has become the focus of research.Vehicle safety is divided into active safety and passive safety,active safety is to carry out early warning before the accident,and passive safety is to reduce the damage caused by the accident after the accident.The forward vehicle collision warning system is an important part of active safety technology.When the vehicle is about to hit the front vehicle,the system can send out sound to remind the driver,which can effectively reduce the occurrence of traffic accidents.This paper aims to research the forward vehicle collision warning system based on iOS platform,access road image information using a mobile phone camera,and then process the acquired image,when it is judged that the vehicle has the risk of collision with the front vehicle,the system will remind the driver by sounding a warning.In this paper,the design of forward vehicle collision warning system is divided into four parts: vehicle detection,vehicle tracking,lane detection and vehicle collision warning.Vehicle detection is the basis of forward vehicle collision warning system.In order to ensure the real-time detection,this paper uses the Haar-like features and Adaboost classifier to detect the vehicle.Meanwhile,we adopt the active learning to further optimize the performance of the classifier.In the detection process,maybe due to changes in light and other reasons,which can casuse the same vehicle in the previous few frames was detected but in the current frame is not detected,so this paper introduces the Kalman filter tracking,predict the location of vehicle in the current frame according to the vehicle's locations in previous frame.The system warns the driver when the front vehicle located in the same lane with the current vehicle.Therefore,it is necessary to judge whether the front vehicle is in the same lane with the current vehicle.In this paper,we use the edge-based lane line detection method to detect the lane line,after detecting the lane line position,take the front vehicle middle bottom point as the vehicle's location,and according to the geometric relationship to determine whether the front vehicle is in the same lane with the current vehicle.For vehicle collision warning,we adopt a method to calculate the collision time according to the width change of the front vehicle in image plane.In this paper,the algorithm is tested on the PC platform and iOS platform,and the performance in each aspect have achieved satisfactory results.