On The Development Of Single Image Based On Positioning And Orienting System

Positioning system is ubiquitous in modern life.Since the appearance of satellite navigation and positioning technology represented by GPS,it has been widely used in the world due to its excellent properties such as wide coverage,no influence of climate and automatic operation.With China's Beidou system put into civilian use and the formal provision of regional positioning and navigation services,the use of navigation satellites for positioning and navigation has basically met people's needs for outdoor positioning and navigation services.In the indoor environment,individual users and intelligent vehicles also have urgent needs for positioning.However,in the complex building environment of the city,due to the shelter of the building,the stability of the satellite navigation signal received by the navigated object is affected,which leads to the decrease of the positioning accuracy of the navigation satellite indoors.In order to make up for this defect,various indoor positioning technologies have emerged as the times require.Currently,commonly used positioning technologies include radio frequency,Bluetooth,Wi-Fi,infrared,ultrasonic and computer vision-based positioning methods.In the process of industrial production,in order not to be affected by the production environment,the positioning method using computer vision is often selected.Therefore,it is of great significance to develop an indoor positioning and attitude determination system which can be applied in industrial environment and can realize fast and high precision.Using the geometric invariants of image points and their corresponding points in the spatial field,the matching of image points and spatial points is realized.The method of camera resection is studied,and a resection algorithm using homography matrix is proposed.The hardware system and software system of the positioning and attitude determination sensor are set up to complete the development of the positioning and attitude determination sensor.The main research contents and methods are as follows:1.This paper studies the invariants between the five points of the conplanar projection in three-dimensional space and the corresponding five points of the two-dimensional image in perspective projection transformation,and uses the invariants to realize the matching and correspondence between the three-dimensional space points and the two-dimensional image points.Using the geometric invariants of projection projection,the corresponding relation between spatial points and image points is obtained.The matching of spatial points and image points is completed.Using cross ratio invariants,coding points can be designed,and matching of homonymous image points and control field points can be completed through identification of coding points.2.In this paper,a new algorithm of posterior resection is proposed.Byestablishing a temporary coordinate system of three points in space,the algorithm transforms the solution of the posterior resection into the solution of a ternary quadratic system of equations by taking advantage of the fact that the coordinate components of the three points in the temporary coordinate system are 0,and the solution of the equations is solved by Newton iteration method.3.The hardware platform of positioning and positioning sensor system is designed and the software of matching and positioning and positioning algorithm between 3d space point and 2d image point is developed.