Image matching type indoor positioning method with assistance of crowd sensing WiFi signal fingerprint

Abstract

The invention provides an image matching type indoor positioning method with the assistance of a crowd sensing WiFi signal fingerprint. According to the method, preliminary positioning is realized by means of WiFi; and according to a preliminary positioning result, auxiliary image positioning is carried out. Therefore, image positioning complexity is reduced; and the real-time performance of the indoor positioning result is improved.

Description

technical field [0001] The invention relates to the technical field of indoor positioning, in particular to a crowd-sensing WiFi signal fingerprint-assisted image matching indoor positioning method. Background technique [0002] Location information is an important basis for location-based service applications. GPS (Global Positioning System) or GNSS (Global Navigation Satellite System) is often used to obtain target location data for outdoor environment positioning. GPS refers to the use of GPS positioning satellites to perform real-time global positioning. Positioning and navigation systems. However, there is an obvious defect in GPS: it cannot be positioned indoors, and the accuracy of civilian GPS is usually only about 10m, which is still a distance away from the indoor navigation requirement (about 1m). Moreover, in the indoor environment, due to GNSS signal attenuation, occlusion and other reasons, the information of the indoor environment cannot be obtained, which ma...

AI Technical Summary

Problems solved by technology

However, GPS has an obvious defect: it cannot locate indoors, and the accuracy of civilian GPS is only about 10m, which is still a long way from the indoor navigation requirements (about 1m).

[0006] There are other problems in using this reference point database acquisition method: for some areas, due to indoor layout and other reasons, it is impossible to obtain high-density reference points; secondly, the cost of maintaining a database is also very huge

Method 1) needs to use the theory of photogrammetry, usually the calculation process is more complicated, and requires special and high-precision measuring instruments for the positioning process, which will increase the cost of realizing positioning. Method 2) needs to use image Matching recognition realizes indoor positioning. With the development of optical sensing technology and image recognition technology and the improvement of smartphone processing capabilities, the positioning accuracy can be greatly improved. However, the positioning system based on image matching needs to build a huge image database. In the real-time matching stage, the images taken by the user need to be searched and queried in the entire database to find matching images. This process takes a long time, resulting in poor real-time performance of the positioning method as a whole.

Images