A SIFT Feature Matching Algorithm Based on Semi-Variance Function

2013 ◽  
Vol 647 ◽  
pp. 896-900 ◽  
Author(s):  
Feng Tian ◽  
Yu Bo Yan
Keyword(s):  
High Dimension ◽  
Image Matching ◽  
Feature Matching ◽  
Variance Function ◽  
Feature Point ◽  
Matching Algorithm ◽  
Sift Feature ◽  
Matching Efficiency

For solving the low matching efficiency problem due to high dimension of eigenvector in SIFT, a SIFT feature matching algorithm based on semi-variance function is proposed. For each feature point in image SIFT feature point zone, m beams are generated by using the position of the feature point as center and the orientation of the feature point as start direction. The image semi-variance function value of each beam, which is treated as SIFT value of eigenvector descriptor, is used in the algorithm aiming at reducing the dimension of eigenvector and improving image matching efficiency. The experiment result shows that the matching rate of this algorithm is higher, the matching time of this algorithm is less.

Reference Point-Based SIFT Feature Matching

2014 ◽  
Vol 543-547 ◽  
pp. 2670-2673
Author(s):  
Lei Cao ◽  
Di Liao ◽  
Bin Dang Xue
Keyword(s):  
Reference Point ◽  
Image Matching ◽  
Euclidean Distance ◽  
Feature Matching ◽  
Experimental Results ◽  
Feature Descriptor ◽  
Matching Algorithm ◽  
Sift Feature ◽  
Sift Features ◽  
Suitable Reference

Aiming to solve the high computational and time consuming problem in SIFT feature matching, this paper presents an improved SIFT feature matching algorithm based on reference point. The algorithm starts from selecting a suitable reference point in the feature descriptor space when SIFT features are extracted. In the feature matching stage, this paper uses the Euclidean distance between descriptor vectors of the feature point to be matched and the reference point to make a fast filtration which removes most of the features that could not be matched. For the remaining SIFT features, Best-bin-first (BBF) algrithm is utilized to obtain precise matches. Experimental results demonstrate that the proposed matching algorithm achieves good effectiveness in image matching, and takes only about 60 percent of the time that the traditional matching algorithm takes.

Research on the SIFT Algorithm in Image Matching

2011 ◽  
Vol 121-126 ◽  
pp. 4656-4660 ◽  
Author(s):  
Yuan Cong ◽  
Xiao Rong Chen ◽  
Yi Ting Li
Keyword(s):  
Image Matching ◽  
Feature Matching ◽  
Scale Space ◽  
Feature Point ◽  
Image Feature ◽  
Sift Algorithm ◽  
Matching Algorithm ◽  
Key Points ◽  
Point Neighborhood ◽  
Accurate Matching

SIFT feature matching algorithm is hot in the field of the currently feature matching research, its matching with the strong ability can deal with the translation, rotation, affine transformation occurring between images , and it also have a stable image feature matching ability to the images filmed at any angle. SIFT algorithm is adopted in this paper, matching feature point through the scale space , calculating the histogram of detecting feature point neighborhood of gradient direction characteristic vector and generation to SIFT eigenvector and the key points similarity measure. From different setting threshold, scale scaling, rotating, noise on the experiment, the experiment result proves this algorithm in the above aspects has good robustness, suitable for mass characteristic database of rapid, accurate matching.

The Optimization of SIFT Feature Matching Algorithm on Face Recognition Based on BP Neural Network

2015 ◽  
Vol 743 ◽  
pp. 359-364 ◽  
Author(s):  
B. Liao ◽  
H.F. Wang
Keyword(s):  
Neural Network ◽  
Bp Neural Network ◽  
Feature Matching ◽  
Illumination Variation ◽  
Sift Algorithm ◽  
Matching Algorithm ◽  
Sift Feature ◽  
Identification Rate ◽  
Candidate Image ◽  
Matching Efficiency

In the field of object recognition, the SIFT feature is known to be a very successful local invariant descriptor and has wide application in different domains. However it also has some limitations, for example, in the case of facial illumination variation or under large tilt angle, the identification rate of the SIFT algorithm drops quickly. In order to reduce the probability of mismatching pairs, and improve the matching efficiency of SIFT algorithm, this paper proposes a novel feature matching algorithm. The basic idea is taking the successful-matched SIFT feature points as the training samples to establish a space mapping model based on BP neural network. Then, with the help of this model, the estimated coordinate of the corresponding SIFT feature point in the candidate image is predicted. Finally search the possible matching points around the coordinate. The experiment results show that using the prediction model, the number of mismatching points can be reduced effectively and the number of correct matching pairs increases at the same time

Image Feature Matching Based on SIFT Algorithm

2014 ◽  
Vol 644-650 ◽  
pp. 4157-4161
Author(s):  
Xin Zhang ◽  
Ya Sheng Zhang ◽  
Hong Yao
Keyword(s):  
Image Matching ◽  
Feature Matching ◽  
Quadratic Function ◽  
Scale Space ◽  
Feature Point ◽  
Image Feature ◽  
Three Dimension ◽  
Feature Points ◽  
Gradient Distribution ◽  
Sift Feature

In the process of image matching, it is involved such as image rotation, scale zooming, brightness change and other problems. In order to improve the precision of image matching, image matching algorithm based on SIFT feature point is proposed. First, the method of generating scale space is introduced. Then, the scale and position of feature points are determined through three dimension quadratic function and feature vectors are determined through gradient distribution characteristic of neighborhood pixels of feature points. Finally, feature matching is completed based on the Euclidean distance. The experiment result indicates that using SIFT feature point can achieve image matching effectively.

scholarly journals SIFT Feature Matching Algorithm with Local Shape Context

2013 ◽  
Vol 5 (20) ◽  
pp. 4810-4815
Author(s):  
Gu Lichuan ◽  
Qiao Yulong ◽  
Cao Mengru ◽  
Guo Qingyan
Keyword(s):  
Feature Matching ◽  
Shape Context ◽  
Local Shape ◽  
Matching Algorithm ◽  
Sift Feature

An Image Mosaic Method Based on SIFT Feature Matching

2012 ◽  
Vol 433-440 ◽  
pp. 5420-5424 ◽  
Author(s):  
Li Jing Cao ◽  
Ming Lv
Keyword(s):  
Image Matching ◽  
Feature Matching ◽  
Image Mosaic ◽  
Sift Feature ◽  
Mosaic Image ◽  
Blending Method ◽  
Orientation Scale ◽  
Sift Features ◽  
Human Eyes ◽  
Arbitrary Matching

This paper concerns the problem of image mosaic. An image matching method based on SIFT features and an image blending method of improved Hat function are proposed in the paper. SIFT feature is local feature and keeps invariant to scale zoom, rotation and illumination. It is also insensitive to noise, view point changing and so on. Because of this our method is insensitive to orientation, scale and illumination of input images, so it’s possible to accomplish image mosaic between arbitrary matching images and the Hat function blending algorithm with global intensity revise makes the mosaic image accepted by human eyes.

An Improved SIFT Matching Algorithm

2012 ◽  
Vol 239-240 ◽  
pp. 1232-1237 ◽  
Author(s):  
Can Ding ◽  
Chang Wen Qu ◽  
Feng Su
Keyword(s):  
High Dimension ◽  
Gradient Method ◽  
Feature Matching ◽  
Feature Descriptor ◽  
Scale Invariant ◽  
Matching Algorithm ◽  
Invariant Feature ◽  
Feature Transform ◽  
New Feature ◽  
Scale Invariant Feature

The high dimension and complexity of feature descriptor of Scale Invariant Feature Transform (SIFT), not only occupy the memory spaces, but also influence the speed of feature matching. We adopt the statistic feature point’s neighbor gradient method, the local statistic area is constructed by 8 concentric square ring feature of points-centered, compute gradient of these pixels, and statistic gradient accumulated value of 8 directions, and then descending sort them, at last normalize them. The new feature descriptor descend dimension of feature from 128 to 64, the proposed method can improve matching speed and keep matching precision at the same time.

SIFT feature matching algorithm of multi-source remote image

2013 ◽  
Vol 21 (8) ◽  
pp. 2146-2153 ◽  
Author(s):  
刘志文 LIU Zhi-wen ◽  
刘定生 LIU Ding-sheng ◽  
刘鹏 LIU Peng
Keyword(s):  
Feature Matching ◽  
Matching Algorithm ◽  
Sift Feature

scholarly journals A NEW PARADIGM FOR MATCHING UAV- AND AERIAL IMAGES

2016 ◽  
Vol III-3 ◽  
pp. 83-90 ◽  
Author(s):  
T. Koch ◽  
X. Zhuo ◽  
P. Reinartz ◽  
F. Fraundorfer
Keyword(s):  
Real World ◽  
Image Matching ◽  
Feature Matching ◽  
Ground Truth ◽  
Geometric Constraints ◽  
Feature Point ◽  
Aerial Images ◽  
Aerial Image ◽  
Ratio Test ◽  
New Paradigm

This paper investigates the performance of SIFT-based image matching regarding large differences in image scaling and rotation, as this is usually the case when trying to match images captured from UAVs and airplanes. This task represents an essential step for image registration and 3d-reconstruction applications. Various real world examples presented in this paper show that SIFT, as well as A-SIFT perform poorly or even fail in this matching scenario. Even if the scale difference in the images is known and eliminated beforehand, the matching performance suffers from too few feature point detections, ambiguous feature point orientations and rejection of many correct matches when applying the ratio-test afterwards. Therefore, a new feature matching method is provided that overcomes these problems and offers thousands of matches by a novel feature point detection strategy, applying a one-to-many matching scheme and substitute the ratio-test by adding geometric constraints to achieve geometric correct matches at repetitive image regions. This method is designed for matching almost nadir-directed images with low scene depth, as this is typical in UAV and aerial image matching scenarios. We tested the proposed method on different real world image pairs. While standard SIFT failed for most of the datasets, plenty of geometrical correct matches could be found using our approach. Comparing the estimated fundamental matrices and homographies with ground-truth solutions, mean errors of few pixels can be achieved.

Sign in / Sign up

Export Citation Format

Share Document