| Autor: |
Zhiyang Liu, Seokwon Chang |
| Jazyk: |
angličtina |
| Rok vydání: |
2024 |
| Předmět: |
|
| Zdroj: |
Measurement: Sensors, Vol 31, Iss , Pp 100970- (2024) |
| Druh dokumentu: |
article |
| ISSN: |
2665-9174 |
| DOI: |
10.1016/j.measen.2023.100970 |
| Popis: |
This paper addresses the challenge of enhancing the presentation style of visual exhibitions within museum settings, where the traditional approach can be perceived as unengaging. The study focuses on improving the performance of the ORB algorithm, a feature detection and matching method. Specifically, we introduce a novel technique that incorporates scale invariance into the algorithm by borrowing the detection concept from the BRISK algorithm. Furthermore, we integrate an improved PROSAC algorithm into the matching stage to enhance the quality of matched points. Our experimental results demonstrate significant improvements in the accuracy of the ORB algorithm. In comparison to the SIFT and GMS algorithms, the ORB algorithm exhibits enhanced matching accuracy, making it a compelling choice for feature-based computer vision applications. Moreover, the proposed approach creates a captivating illusion of seamlessly merging reality and virtual elements, meeting the expectations of system designers. This integration enhances the overall visitor experience within museum settings, fostering engagement and interaction. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
