Bayesian regularization back propagation neural network coordinate conversion method and device

Abstract

The invention discloses a Bayesian regularization back propagation neural network coordinate conversion method and device, which are used for solving the problems that in the coordinate conversion process, a set of coordinate conversion parameters is adopted in the whole area, so that the precision of the local area is lost in the coordinate conversion process, and the accuracy of the converted coordinates is influenced. The Bayesian regularization back propagation neural network coordinate conversion method comprises the following steps: determining plane coordinates of a 2000 country geodetic coordinate system and plane coordinates of an engineering independent coordinate system of a control point, and creating a learning set; training a neural network model according to the learning setuntil the performance index of the neural network model reaches a preset value, and determining that the training is completed; and according to the trained neural network model and the 2000 countrygeodetic coordinate system plane coordinates of the to-be-measured point, determining engineering independent coordinate system plane coordinates of the to-be-measured point obtained through conversion. According to the Bayesian regularization back propagation neural network coordinate conversion method, the network weight of the back propagation neural network is limited through the Bayesian regularization algorithm, and the network structure is effectively improved, and the accuracy of coordinate conversion is improved.

Description

technical field [0001] The present application relates to the technical field of coordinate conversion, in particular to a method and device for coordinate conversion of a Bayesian regularized backpropagation neural network. Background technique [0002] Real-time Kinematic (RTK) is a differential positioning technology based on GPS carrier phase observations. It can provide centimeter-level, real-time positioning methods and has a wide range of applications. [0003] In engineering applications, based on the geodetic coordinates obtained by RTK technology, the coordinates of the 2000 national geodetic coordinate system (CGCS2000) can be obtained through Gaussian projection for application in engineering construction. However, in many major projects, in order to avoid errors caused by Gaussian projection deformation, the coordinates of the national geodetic coordinate system are usually converted into independent engineering coordinates for practical engineering applications...

AI Technical Summary

Problems solved by technology

[0006] The embodiment of the present application provides a Bayesian regularized backpropagation neural network coordinate conversion method and device to solve the problem that in the process of coordinate conversion, if a set of coordinate conversion parameters is used for the entire area, the accuracy of the local area will be lost in the coordinate conversion. Issues affecting the accuracy of transformed coordinates

Images