基于背景轨迹筛选的电子后视镜稳像算法  

Electronic Rearview Mirror Image Stabilization Algorithm Based on Background Trajectory Screening

作　　者：杨朝阳[1] 黄健强 YANG Zhaoyang;HUANG Jianqiang(Hubei University of Automotive Technology,Shiyan,Hubei 442002,China)

机构地区：[1]湖北汽车工业学院,湖北十堰442002

出　　处：《汽车零部件》2024年第8期1-5,共5页Automobile Parts

基　　金：湖北省高等学校优秀中青年科技创新团队T201913;机-电-磁-液-热耦合系统的一体化瞬态数学建模与计算方法研究2021CFB592。

摘　　要：车辆行驶过程中,电子后视镜图像常出现抖动现象,影响驾驶安全性和舒适性。传统稳像算法在处理快速运动的场景时,由于受到视差的影响,难以对摄像头运动进行准确估计,导致稳像效果显著下降。为了解决这一问题,提出了基于特征轨迹模长与仿射变换映射误差的轨迹筛选方法,将特征轨迹分为背景轨迹与前景轨迹。之后采用卡尔曼滤波算法对能够反映真实运动的背景轨迹运动参数进行平滑处理,提升稳像效果。试验结果表明,所提出的电子后视镜稳像算法可以有效提高电子后视镜图像稳定性。During vehicle operation,the electronic rearview mirror image often shakes,which affects driving safety and comfort.Traditional image stabilization methods are difficult to accurately estimate camera motion in fast-moving scenes due to the influence of disparity,resulting in a significant decrease in image stabilization effect.To address this issue,a trajectory screening method based on feature trajectory module length and affine transformation mapping error is proposed,which divides feature trajectories into background trajectories and foreground trajectories.Afterwards,the Kalman filtering algorithm is used to smooth the motion parameters of the background trajectory that can reflect real motion,achieving the effect of image stabilization.The experimental results show that this method can effectively improve the stability of electronic rearview mirror images.

关 键 词：电子后视镜 视频稳像 特征轨迹 

分 类 号：TH122[机械工程—机械设计及理论]