Title :
A Method of Image Matching Used in Image-Based Modeling System
Author :
Yi, Chengtao ; Wang, Xiaotong ; Xu, Xiaogang
Author_Institution :
Dalian Naval Acad., Dalian, China
Abstract :
In order to restrain the high sensitivity to image noise and non-linearity transform as for the traditional automatic matching algorithm in the system of image-based modeling, a new simplified algorithm based on SIFT (scale invariant feature transform) was provided. Firstly, for avoiding the problem of losing of information, position excursion and the fake keypoints, the features were detected and captured in multi-scale space. Secondly, the reversible image matching algorithm was adopted based on simplifying SIFT local feature descriptor for accurate matching. Lastly, the matching algorithm was optimized by using RANSAC and the approximate nearest neighbor algorithm in the light of epipolar constraints. The experimental results demonstrated the robustness and efficiency of the algorithm.
Keywords :
feature extraction; image matching; object detection; random processes; RANSAC; SIFT local feature descriptor; approximate nearest neighbor algorithm; epipolar constraint; feature detection; image matching; image noise; image-based modeling system; multiscale space; nonlinearity transform; scale invariant feature transform; Artificial intelligence; Change detection algorithms; Computational intelligence; Computer vision; Data mining; Feature extraction; Image matching; Image reconstruction; Layout; Stability; Epipolar; RANSAC; SIFT; image matching;
Conference_Titel :
Artificial Intelligence and Computational Intelligence, 2009. AICI '09. International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-3835-8
Electronic_ISBN :
978-0-7695-3816-7
DOI :
10.1109/AICI.2009.32
