Logarithmic transformation for enhancing keypoint matching of SIFT in augmented reality

Augmented Reality (AR) is the technology to combine virtual information and the real world. Scale-Invariant Feature Transform (SIFT) is a method that could be implemented in AR. SIFT has good stability and invariance, give high performance, and implemented by many applications. However SIFT itself h...

Full description

Saved in:
Bibliographic Details
Main Authors: Pambudi, Elindra Ambar, Fauzan, Achmad, B., Abid Yanuar, Sugiyanto, Sigit
Format: Conference Proceeding
Language:English
Subjects:
Algorithms
Augmented reality
Image processing
Image quality
Logarithms
Matching
Rotation
Transformations
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Augmented Reality (AR) is the technology to combine virtual information and the real world. Scale-Invariant Feature Transform (SIFT) is a method that could be implemented in AR. SIFT has good stability and invariance, give high performance, and implemented by many applications. However SIFT itself has a disadvantage, it is not the flawless matching of the SIFT descriptor. In this research, we propose one of the image processing techniques, which is called logarithmic transformation. Our research aims to produce output value of combination Logarithmic Transformation and SIFT algorithm. The steps of our proposed method start from extracting video to sequential image, increase the quality of each frame using logarithmic transformation, and implement SIFT and evaluate this method with total keypoint matching. Combination LIP and SIFT will be compared with SIFT standard. We examined our research with rotation and distance. The result of LIP SIFT can increase the performance of SIFT standard as much as 16% for rotation and 1,5 % for distance-based SIFT standard.
ISSN:0094-243X
1551-7616
DOI:10.1063/5.0106221