Font Size: a A A

The Study Of Indoor Location Algorithm Based On Image

Posted on:2018-05-25Degree:MasterType:Thesis
Country:ChinaCandidate:X Q WangFull Text:PDF
GTID:2348330533462791Subject:Geodesy and Survey Engineering
Abstract/Summary:PDF Full Text Request
In the era of big data,with the rapid development of Internet technology and intelligent mobile phone,mobile services based on location which plays an important role in various industries and people's daily life.According to statistics,most people spend 80%of their time indoors,High precision of indoor positioning can bring convenience to people's work and life,especially in museums,cinemas and shopping malls.Aiming at the problems of low position-ing accuracy,high cost and poor portability in traditional indoor positioning methods.In this paper,a method of spatial location based on image in the C/S(Client-Server)architecture is proposed.This method has many advantages,such as high precision,good visualization effect,abundant information and portability,which makes it become one of the research hotspots in the field of indoor positioning in recent years.This paper mainly through the client and server two parts to achieve indoor positioning.In the client to achieve the function is to call the camera to capture the image and upload the image to server.On the server,the feature library is established in advance and when the location image is uploaded successfully,the exterior orientation elements are calculated and when the calculating is complete,the exterior orientation elements will be send back to the client.Finally,we show the user location in 3D scene.The content and innovation of this paper are listed as follows:(1)The pinhole imaging model and the camera model based on collinearity condition equation are introduced in detail.The advantages and disadvantages of the existing camera calibration methods are analyzed.The 3D camera calibration control field of Xi'an University of Science and Technology is used to obtain the parameters of camera and lens distortion parameters.(2)The SFM(structure from motion)method is used to calculate the coordinates of the object points.The feature points of the image sequence and the coordinates of the image points,and the 3D coordinates of the object points are arranged into the feature database of and the object point coordinates are provided for calculating the exterior orientation elements of the positioning image.(3)The exterior orientation elements of the positioning image are calculated.The feature points of location image are extracted and matched with the feature point of object database to get the object space coordinates.The projection matrix H is obtained by the method of maximum likelihood estimation,and then the projection matrix is decomposed to obtain the initial orientation of the image.Finally,the exterior orientation elements of the image can be obtained by the collinearity equation.The experimental results show that the proposed method can achieve centimeter level of positioning accuracy and can meet the users'requirements.(4)The 3D scene model and the related code are packaged into the mobile phone software to realize the function of calling the camera to capture the positioning image and the three-dimensional roaming display positioning result.
Keywords/Search Tags:indoor localization, image matching, three-dimensional reconstruction, feature library of object space, exterior orientation elements
PDF Full Text Request
Related items