An empirical study of some feature matching strategies

Pattern Recognition Letters, 1998

A two step feature matching algorithm which is primarily aimed at problems related to the analysis of aerial images of man-made sites is presented. Only linear features and their geometric attributes are used in the algorithm. First, the rotation between the two images is calculated and then matching by relaxation is performed assuming that there is only translation.

This paper presents a statistical method to match feature points from stereo pairs of images. The proposed method is evaluated in terms of effectiveness, robustness and computational speed. The evaluation was performed on several pairs of real stereo images of natural scenes taken onboard an unmanned aerial vehicle. The results show that the proposed method reduces the number of incorrect matches and is fast. Cet article décrit une méthode de mise en correspondance de points d’intérêts extraits d’images stéréoscopiques. Cette méthode a été évaluée en termes d’efficacité, de robustesse et de temps de calcul. L’évaluation a porté sur plusieurs paires d’images prises dans un environnement naturel à partir d’un banc stéréoscopique embarqué sur un drone d’intérieur. Les résultats montrent que la méthode proposée est très rapide et réduit considérablement le nombre de mauvais appariements. 1.

2010 Canadian Conference on Computer and Robot Vision, 2010

In this report we introduce a novel approach for determining correspondence in a sequence of images. We formulate a probabilistic framework that relates a feature's appearance and its position under relaxed statistical assumptions. We employ a Monte-Carlo approximation for the joint probability density of the feature position and its appearance that uses a flexible noise and motion model to generate random samples. The joint probability density is modeled by a Gaussian Mixture. The feature's position given its appearance is then determined by maximizing its posterior. We evaluate our method using real and synthetic sequences and compare its performance with leading or popular algorithms from the literature. The noise robustness of our algorithm is superior under a wide variety of conditions. The method can be applied in the context of optical flow, tracking and any application that needs feature point matching.

Image matching is the process of bringing two images geometrically into agreement so that matching pixels in the two images matches to the same physical region of the area being imaged. Matching algorithms plays a key role in deciding correspondences between two image scenes. The matching algorithms are of two types area based matching and feature based matching. This paper aims to present a review of various image matching methods.

Image matching refers to the automatic identification and measurement of corresponding image points that are located on the overlapping area of multiple images. Matching algorithms play an important role in deciding correspondences between two image scenes. The proficiency of an image matching algorithm depends on the region of experimental set up and assumptions considered by the researchers. Several image matching techniques have been studied to address this problem. This study highlights latest methodologies and feature extraction methods used by researchers for image matching.

1995

Image matching is widely considered to be one of the most difficult tasks of a digital photogrammetric system. Traditionally image matching has been approached from either an area based or a feature based point of view. In recent years significant progress has been made in Area Based Matching (ABM) techniques such as Multiphoto Geometrically Constrained Least Squares Matching. Also in the field of Feature Based Matching (FBM) improvements have been made in extracting and matching image features, using for example the Forstner Operator followed by feature matching. Generally, areaand feature based matching techniques have been developed independently from each other. The aim of this research project was to design an automated image matching scheme that combines aspects of Feature Based Matching (FBM) and Area Based Matching (ABM). The reason for taking a hybrid approach is to encapsulate only the advantages of each matching scheme while cancelling out the disadvantages. The approach ...

2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07, 2007

Most existing feature-matching methods utilize texture correlation for feature matching, which is usually sensitive to contrast changes. This paper proposes a new feature-point matching algorithm that does not rely on the image texture. Instead, only the smoothness assumption, which states that the displacement eld in a neighborhood is coherent (smooth), is used. In the proposed method, the collected correspondences of a group of feature points within a neighborhood are ef ciently determined such that the coherence measure of the displacement eld in the neighborhood is maximized. The experimental results show that the proposed method is invariant to contrast changes and signi cantly outperforms the conventional blockmatching technique.

Educational Research Review, 2003

Two problems must be solved in order to estimate the motion of an un-calibrated camera. One is the determination of the camera's parameters and the other is the correspondence matching of some features in images taken at various locations. In this paper, we propose a new technique to extract corners automatically in the sequence of images and then match the

International Journal of Environment, 2019

Features are distinctive landmarks of an image. There are various feature detection and description algorithms. Many computer vision algorithms require matching of features from two images. Large number of correct matches with homogeneous distribution in the images is needed for robustness of the image matching. The matches are generally obtained using a feature distance threshold and ambiguous matches are rejected using a ratio test. This paper proposes a method that can be added to image matching pipeline for enhancing homogeneous distribution and increasing the number of matched feature points. After successfully matching an image pair, spatially close feature points go through an elimination process which aims to decrease ambiguity at the second matching step. Then, a coarse geometric transformation between two images is calculated, through which the detected feature points in one image (i.e. the moving image) are projected to the other image (i.e. the fixed image). Then, featur...

International Journal of …, 2010

Image matching is an important task to be performed for the correspondence problem. The distinction between different matching primitives is probably the most prominent difference between the various matching algorithms. The primitives fall into two broad categories: either windows composed of grey values or features extracted in each image a priori are used in the actual matching step. The resulting algorithm usually called: Area Based Matching (ABM), and Feature Based Matching (FBM) respectively. In this paper, the comparative study of these two algorithms has been done which results with one best optimal method i.e. Feature Based Matching. FBM procedures widely used in pattern recognition and computer vision & fine increasing interest also in photogrammetry.