Augmented Reality Tourism Technology Based on an Improved ORB Algorithm and Homography Matrix

Area of Science:

• Computer Vision
• Augmented Reality
• Tourism Technology

Background:

• Augmented reality (AR) tourism applications face challenges with feature-point matching and 3D registration.
• Existing methods suffer from poor alignment, low stability in complex environments, and inaccurate feature matching.

Purpose of the Study:

• To develop an improved feature-point matching and 3D registration method for AR tourism.
• To enhance alignment accuracy, stability, and feature-matching precision in AR applications.

Main Methods:

• Combined Speeded Up Robust Features (SURF) with enhanced Oriented FAST and Rotated BRIEF (ORB) for feature detection and matching.
• Employed Lucas-Kanade optical flow for tracking and Random Sample Consensus (RANSAC) for outlier elimination.
• Proposed an augmented reality tourism 3D registration technique using an improved homography matrix.

Main Results:

• Achieved average correct matching rate increases of 44.08% (vs. SIFT), 36.51% (vs. SURF), and 16.09% (vs. original ORB) under scale variation.
• Demonstrated improved correct matching rates under fuzzy transformations (33.46%, 19.65%, 9.35% increases).
• Attained an average registration accuracy of 98.74% under rotational transformation.

Conclusions:

• The proposed method significantly enhances feature-point matching and 3D registration for AR tourism.
• The improved technique offers a more stable and accurate solution for virtual-real world fusion in travel applications.
• This research provides a novel approach for advancing AR technology in the tourism industry.