Research And Application Of Video Stabilization Technology For Vehicle Electronic Rearview Mirror

With the rapid economic development and technological progress,electronic rearview mirrors have replaced traditional glass mirrors and become the current development trend in terms of solving the lack of vehicle vision and saving fuel resources.In practical applications,due to problems such as wind excitation and road bumps,the video displayed by the on-board electronic rear-view mirror system(hereinafter referred to as the on-board video)jitters.Observing for a long time not only increases the driver’s visual fatigue,but also increases the risk when driving.In order to eliminate the impact of jitter on vehicle video and improve driving safety,an urgent need for vehicle video image stabilization technology is put forward.This thesis takes the on-board electronic rearview mirror system as the main research object.According to the jitter characteristics of the vehicle under different motion states,the key technology of on-board video image stabilization is deeply studied.The specific research content is as follows:(1)Vehicle-mounted video image stabilization algorithm based on improved image fusion splicingWhen the vehicle is idling,the in-vehicle video has the characteristics of small viewing angle translation and small angle rotation jitter in a fixed scene.Aiming at the traditional image stabilization algorithm,when dealing with the above problems,the image stabilization effect is mediocre and the real-time performance is poor.This thesis proposes a vehicle-mounted video image stabilization algorithm based on improved image fusion splicing.First,extract the high-information video frame sequence through the sampling process of the vehicle-mounted video;then use the improved image fusion splicing technology to obtain the image of the wide viewing angle area,and finally use the cropping method to obtain the stabilized video frame sequence.(2)Vehicle-mounted video image stabilization algorithm based on different weighted motion estimationWhen the vehicle is driving,the on-board video has the characteristics of complex background motion and uneven illumination.For traditional electronic image stabilization algorithms,when performing real-time image stabilization,it is difficult to extract image feature points,showing poor image stabilization robustness and real-time image stabilization problems.This thesis proposes a vehicle-mounted video image stabilization algorithm based on different weighted motion estimation.In improving the matching accuracy and efficiency of feature points,the correct matching point pairs are extracted through the partitioned feature point extraction strategy and the optimized RANSAC algorithm.Then,the motion estimation method with different weights and the affine transformation model are used to estimate the motion parameters,and the image frames are reversely compensated to obtain a stable video frame sequence.(3)Application of Video Stabilization Technology of Electronic Rearview MirrorIn order to solve the problem of insufficient vehicle vision,this thesis designs a vehicle-mounted electronic rearview mirror system.Through the test experiment on the visual field coverage of the electronic sight glass,it is verified that the system has strong effectiveness and practicability in solving the problem of insufficient vehicle visual field.Aiming at the problem of vehicle video jitter in the actual application of the vehicle electronic rearview mirror system,combined with the vehicle image stabilization algorithm proposed in this thesis,through the vehicle video real-time image stabilization test experiment,it is verified in the comparison test with the traditional video image stabilization algorithm The video image stabilization algorithm proposed in this thesis has good robustness and real-time performance.This method has been successfully applied in the actual project of video image stabilization of electronic rearview mirrors.