Indoor landmark updating method and device based on evidence fusion theory

Abstract

The invention discloses an indoor landmark updating method and device based on an evidence fusion theory, and the method comprises the steps of obtaining a landmark video uploaded by a user, extracting a video sequence frame from the landmark video, and converting the video sequence frame into a video score value; substituting the video score value into a conjecture scene functional expression preset by a user to obtain a conjecture scene probability value; performing first-order fusion calculation and second-order fusion calculation on the conjecture scene probability value to obtain a second-order conjecture scene probability value, and performing second-order data conflict calculation by adopting the conjecture scene probability value to obtain a second-order data conflict index value; when the second-order conjecture scene probability value meets a preset change value requirement and the second-order data conflict index value meets a preset conflict index value, determining that the indoor landmark to be updated reached by the user changes; and replacing the pre-stored sequence frame with the video sequence frame to complete the operation of updating the indoor landmark. According to the invention, the landmark can be automatically updated according to the landmark video, so that the time and cost required for updating are reduced.

Description

technical field [0001] The invention relates to the technical field of indoor positioning, in particular to an indoor landmark update method and device based on evidence fusion theory. Background technique [0002] In recent years, with the rapid development of information technology and the popularization of smart phones, people's production, life and work are increasingly inseparable from various application functions provided by mobile phones. One of the commonly used functions is indoor location positioning. Indoor positioning technology can accurately detect the position of everyone indoors in an emergency, so that rescuers can rescue at the scene. [0003] Traditional indoor positioning technologies mainly include fingerprint matching positioning or video-based indoor positioning. Common methods based on fingerprint matching positioning include Wi-Fi positioning, Bluetooth positioning, RFID positioning and geomagnetic positioning. The above positioning method mainly r...

AI Technical Summary

Problems solved by technology

[0004] The fingerprint matching method for positioning has the following disadvantages. Since the signal needs to be collected point by point in the area to be positioned, and the signal will fluctuate with time, it is necessary to manually resample and update the fingerprint data every once in a while, resulting in slow signal collection. Update efficiency is low and labor costs are too high

However, video-based indoor positioning also requires a lot of manpower and time cost to update and maintain data when indoor landmarks send changes, which also leads to problems of low update efficiency and high labor costs.

Images