Indoor positioning method for smartphones based on hybrid modeling of spatial point-line-surface features

Abstract

An indoor positioning method for smartphones based on hybrid modeling of spatial features, which uses the augmented reality development platform of smartphones to capture and extract various features such as points, lines, and surfaces in the local real scene space, and then uses irregular region segmentation and global coordinate mapping technology Realize the global stitching of multiple local point clouds and 3D space surface line features, generate a global mixed feature model of 3D real scene space, and then match the real-time scanning local space feature set with the global feature model to achieve fast, convenient and high precision Indoor autonomous positioning of smartphones. Based on the real-time positioning results of the mobile phone, by converting the indoor travelable area into a directed graph in advance, the Dijkstra algorithm can be used to find the shortest path and realize indoor navigation.

Description

technical field [0001] The invention relates to the fields of computer vision and indoor positioning and navigation, in particular to an indoor positioning method for smart phones based on hybrid modeling of spatial point, line and surface features. Background technique [0002] Building an accurate spatial model is an essential prerequisite for indoor positioning and navigation based on visual methods. In the past two decades, precise indoor localization and tracking has attracted considerable attention due to its wide application in large indoor environments such as airports, supermarkets, and hospitals. Once accurate location information is obtained in real time, various practical location-based services can be provided, such as indoor navigation services, location information push, and advertising market services. Unfortunately, compared to the GPS technology commonly used in outdoor environments, there is currently a lack of a standard positioning system that can be wi...

AI Technical Summary

Problems solved by technology

These platforms provide advanced application programming interfaces (APIs) for world tracking, space modeling, scene understanding, etc. However, the current indoor autonomous positioning and navigation based on mobile phone platforms has not yet been well resolved.

[0004] The patent application (publication number: CN111508067A) discloses a lightweight indoor modeling method based on vertical planes and vertical lines. Indoor modeling only relying on the vertical planes and vertical lines of the space will result in too little information in the model, making it difficult to Support robust and high-precision indoor positioning

Simultaneous Localization and Mapping (SLAM) is a traditional classic spatial point cloud modeling method, but it has high computational complexity and is difficult to implement on a mobile phone platform with very limited computing resources and storage resources.

More importantly, whether it is a spatial point cloud model or a spatial line-surface model composed of vertical planes and vertical lines, how to quickly and accurately realize indoor positioning based on model matching on the mobile phone platform is still a major technical issue in the field of indoor positioning. problem

Images