Absolute Measurement Method Of Vehicle Panoramic Image Based On LM Optimization
[Purposes]Aiming at the problem of insufficient application of vehicle panoramic images in the fields of smart city,smart city management,smart transportation,etc.,a method of absolute measure-ment of vehicle sequence panoramic images based on LM optimization is proposed.[Methods]Firstly,by analyzing the imaging process of the target point to be measured on the panoramic image,the theoretical model of the absolute coordinate measurement of the target point was deduced in detail.Secondly,in or-der to solve the optimization problem of the theoretical model,the LM algorithm,which is insensitive to the initial value and has a fast convergence speed,is introduced,and the initial value acquisition,func-tion construction and parameter setting in the LM algorithm are designed.Finally,the proposed method is verified by combining the actual vehicle panoramic image and the measured data of the target point.[Findings]The results show that the mean square errors of the plane and elevation coordinates of the tar-get points obtained by this method are 0.123 m and 0.118 m,respectively,which meet the relevant re-quirements in the'Urban Surveying Specification(CJJ/T 8-2011)'.[Conclusions]This method can ef-fectively obtain the absolute coordinates of the target points to be measured in the range of panoramic im-ages,thus providing the possibility for the application of panoramic images in the fields of urban compo-nent survey and transportation infrastructure measurement.
NETL
NSTL
万方数据
