Indoor navigation method based on augmented reality

Abstract

The invention relates to the field of augmented reality and the field of indoor navigation, in particular to an indoor navigation method based on augmented reality, which can be applied to a shopping mall. According to the invention, a shop door face is used as a basis for positioning and navigation, and the system is composed of user-side software and cloud-side software. The user side software is used for acquiring a shop door face image from the camera, transmitting the shop door face image to the cloud side, downloading positioning data from the cloud side and drawing AR navigation information. The cloud software is provided with a deep learning network which completes training and is used for identifying the shop door face. In addition, the cloud software is responsible for editing and storing shopping mall map data and transmitting positioning information. The method comprises the following steps: shooting a shop door face through user side software, transmitting an image to a cloud deep learning network for identification, and returning an identification result; enabling the user side software to retrieve a corresponding shop in the map data according to an identification result returned by the cloud side, and position a user position in a channel in front of a shop door; and enabling a user to input a destination in user side software, and generating AR navigation information according to map data through a current positioning position and a destination position for navigation.

Description

technical field [0001] The invention relates to the fields of augmented reality and indoor navigation, in particular to an augmented reality-based indoor navigation method applicable in shopping malls. Background technique [0002] Unlike outdoors, which can be accurately navigated by GPS, indoor navigation cannot be positioned by GPS. Therefore, indoor navigation is currently mostly positioned by QR codes or Bluetooth devices. Among them, the two-dimensional code positioning method needs to post a large number of two-dimensional codes indoors, which affects the appearance, and the two-dimensional codes are easy to be modified; a large number of bluetooth devices need to be placed for positioning through bluetooth devices, and by detecting the communication signal strength between different bluetooth devices to locate. This method is not only costly, but also the signal strength is easily affected by external interference, resulting in poor accuracy. Contents of the inven...

AI Technical Summary

Problems solved by technology

[0002] Unlike outdoors, which can use GPS for precise navigation, indoor navigation cannot use GPS for positioning. Therefore, indoor navigation is currently mostly based on QR codes or Bluetooth devices.

Among them, the two-dimensional code positioning method needs to post a large number of two-dimensional codes indoors, which affects the appearance, and the two-dimensional codes are easy to be modified; a large number of bluetooth devices need to be placed for positioning through bluetooth devices, and by detecting the communication signal strength between different bluetooth devices locate

This method is not only costly, but also the signal strength is easily affected by external interference, resulting in poor accuracy

Images