A High-Precision Indoor Positioning Method Based on ConvNet Fingerprint Features
By constructing a map of ConvNet fingerprint features and using deep learning for matching, high-precision indoor positioning was achieved, solving the problems of high construction costs and long computation time in existing technologies, and realizing efficient and accurate multi-scale positioning.
CN115862458BActive Publication Date: 2026-05-26HUBEI UNIV OF TECH
View PDF 5 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUBEI UNIV OF TECH
- Filing Date
- 2022-11-29
- Publication Date
- 2026-05-26
Smart Images
Figure CN115862458B_ABST
Abstract
This invention discloses a high-precision indoor positioning method based on ConvNet fingerprint features. First, multiple deep learning features are extracted from the image and constructed hierarchically to form a ConvNet fingerprint feature with strong representational power and high robustness. Then, a high-precision map is constructed based on the ConvNet fingerprint features. Finally, high-precision indoor positioning is achieved by matching deep learning features at different scales. This invention eliminates the need for deploying positioning signal transmitters; high-precision indoor positioning can be achieved solely through images. Compared to existing methods, this method is low-cost and offers high positioning accuracy.
Need to check novelty before this filing date? Find Prior Art