An Indoor Positioning System Based on Static Objects in Large Indoor Scenes by Using Smartphone Cameras.

Area of Science:

• Computer Vision
• Indoor Navigation Systems
• Location-Based Services

Background:

• Increasing demand for location-based services (LBS) in large indoor environments like airports and museums.
• Lack of a universal indoor positioning solution comparable to outdoor Global Navigation Satellite System (GNSS).
• Existing smartphone camera-based methods face limitations in large-scale indoor spaces.

Purpose of the Study:

• To design and develop a novel indoor positioning system for large indoor scenes.
• To enable accurate user location tracking using readily available smartphone cameras.
• To overcome limitations of current visual-based indoor positioning technologies.

Main Methods:

• Utilizing common static indoor objects (e.g., doors, windows) as reference points.
• Integrating deep learning and computer vision algorithms for object detection and smartphone localization.
• Developing a system that requires no additional infrastructure, leveraging existing smartphone hardware.

Main Results:

• The proposed system successfully detects static objects in complex indoor environments.
• Accurate smartphone positioning was achieved, enabling user localization within the indoor space.
• Experimental results in an art museum demonstrated positioning accuracy within 1 meter.

Conclusions:

• The developed system offers a cost-effective solution for indoor positioning in large venues.
• Smartphone cameras combined with deep learning and computer vision provide a viable alternative for indoor navigation.
• The method shows promise for widespread adoption in various LBS applications within large indoor spaces.