Statistical feature point matching method

Proceedings of 13th International Conference on Pattern Recognition, 1996

In this paper, a method of generating structural descriptions of stereo image pairs and their matching to recover 3-D form of objects in space, is discussed. The method consists of the following steps: rstly, straight line segments are extracted in both images and the relations among them are investigated. The relational subgraphs are constructed then, and the correspondence graph is created. Using the "stable marriage" searching algorithm the features in one image corresponding to features in the other image are found. The method was tested on a set of aerial stereo images of urban scenes.

Being as a relatively new approach of signalling, moving-block scheme significantly increases line capacity, especially on congested railways. This paper describes a simulation system for multi-train operation under moving-block signalling scheme. The simulator can be used to calculate minimum headways and safety characteristics under pre-set timetables or headways and different geographic an Abstract Stereo vision is a method of depth perception, in which depth information is inferred from two (or more) images of a scene, taken from di erent perspectives. Applications for stereo vision include aerial photogrammetry, autonomous vehicle guidance, robotics, industrial automation and stereomicroscopy.

Remote Sensing

Image matching is one of the most important tasks in Unmanned Arial Vehicles (UAV) photogrammetry applications. The number and distribution of extracted keypoints play an essential role in the reliability and accuracy of image matching and orientation results. Conventional detectors generally produce too many redundant keypoints. In this paper, we study the effect of applying various information content criteria to keypoint selection tasks. For this reason, the quality measures of entropy, spatial saliency and texture coefficient are used to select keypoints extracted using SIFT, SURF, MSER and BRISK operators. Experiments are conducted using several synthetic and real UAV image pairs. Results show that the keypoint selection methods perform differently based on the applied detector and scene type, but in most cases, the precision of the matching results is improved by an average of 15%. In general, it can be said that applying proper keypoint selection techniques can improve the ac...

[Proceedings] IGARSS'91 Remote Sensing: Global Monitoring for Earth Management, 1991

Area-Based Stereo Matching techniques such as the Adaptive Least-Squares Correlation Region Growing (ALSCRG) method [1,, require an initial set of corresponding points (seed points) in the left and right images to begin stereo matching. This set of initial seed points has, until now, usually been generated manually using a special purpose visual display program. This paper describes three interest operator based methods for automatic generation of seed points for SPOT satellite imagery (class l), aerial photographs digitised on a kern DSR-11 analytical plotter (class 2), and any other imagery with stereo-overlap (class 3.) Results for the three methods and a comparison of interest operators are also presented.

The 9th International Conference on Computing and InformationTechnology (IC2IT2013), 2013

In this paper, we propose a new area-based stereo matching method by improving the classical Census transform. It is a difficult task to match corresponding points in two images taken by stereo cameras, mostly under variant illumination and non-ideal conditions. The classic Census nonparametric transform did some improvements in the accuracy of disparity map in these conditions but it also has some disadvantages. The results were not robust under different illumination, and because of the complexity the performance was not suitable for real-time robotic systems. In order to solve these problems, this paper presents an improved Census transform using Maximum intensity differences of the pixel placed in the center of a defined window and the pixels in the neighborhood to reduce complexity and obtain better performance and needs a smaller window size to obtain best accuracy compared to the Census transform. Experimental results show that the proposed method, achieves better efficiency in term of speed and robustness against illumination changes.

Computers & Geosciences, 2007

Modeling the world in three dimensions has been attracting a growing interest both in applications and science. In many cases, such 3D models are achieved by triangulating corresponding features recorded by several images of the same field taken from different points of view. This, however, requires the ability to match corresponding image elements detected in different images. In this paper, an algorithm for stereo matching in noisy pairs of outdoor images is described. The proposed algorithm applies a standard window-based correlation, but uses a fuzzy logic-based similarity function that models the HSV color space. This fuzzy logic modeling allows a robust color comparison that can tolerate a certain degree of changes in the illumination conditions and can be used for finding corresponding pixels in noisy sets of images. While the proposed algorithm does not introduce an improvement over existing methods in ideal conditions, experiments suggest that it provides significantly better results when the images in the set are relatively different from each other, and can be effective for matching corresponding features in images taken in different weather conditions, different positions of the sun, different optics or other real-life situations in which pinhole conditions are not available.

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.

2020

The process of producing the orthomosaic image can be done in two different ways: image mosaicing and image stitching. In image stitching, there are small overlaps between images in which the images are put together. In contrast, image mosaic requires extended overlaps and a blending of several images. In this study, used images grouped by inclination are also classified according to the type of UAV used. This classification is based on the wing type of the UAS. “Rotary-Wing UAV” and “Fixed-Wing UAV" are selected as two different types at the test field. Images with different characteristics were used to produce mosaics from the algorithms. The first performance preferred a flight height of 30 m (Ground Sample Distance, 0.82 cm/pixel) with the frontal overlap of 80%, and the second performance preferred a flight height of 60 m (GSD, 1.64 cm/pixel) and same overlap. 10 images from both studies were combined in all algorithms. According to the results produced from the mosaics of...

Many camera calibration techniques used in stereoscopic imaging require corresponding points in image pairs to be identified. The best matches tend to come from very distinctive regions, hence it is preferable to use only these points and reject points from more homogeneous regions. Often points are selected and matched by a combination of user input and automated matching. However, for most techniques, the more matching points identified, the more accurate and robust the calibration. Therefore, it is desirable to alleviate the need for user input and to automate the point selection process. In the feature matching technique described here, a combination of object extraction, feature identification and feature matching is used. This technique is simple, efficient and robust. The algorithm will be used as part of a larger project in 3D human motion capture which is currently under development at University College Dublin and the National Rehabilitation Hospital in Dun Laoghaire.

1974

UNCLASSIFIED 15«. DECLASSIFI CATION/DOWN GRADING SCHEDULE 16 DISTRIBUTION STATEMENT 'ol Ihi \ Report) 17 DISTRIBUTION STATEMENT fof the mbetracl entered In Block 30, II dltlerent from Report) 18 SUPPLEMENTARY NOTES PBCK SUBJECT TO QIANGf 19. KEY WORDS (Continue on reveree side il neceeemry and Identity by block number) 20. ABSTRACT fCondnue on reveree tide II neceeeary and Identify by block number; This dissertation describes techniques for efficiently matching corresponding areas of a stereo pair of images. Measures of match which are suitable for this purpose are discussed, as are methods for pruning the search for a match. The mathematics necessary to convert a set ^f matchings into a workable camera model are given, along with calculation, 1 : which us^ this model and a pair of image points to locate the corresponding scene point. Methods are included to detect some types of unmatchable target areas in the original data and for detecting when (contiiWri) DD 1 JAN*73 1473 EDITION OF 1 NOV 65 IS OBSOLETE USßMBSnJM SECURITY CLASSIFICATION OF THIS PAGE flWien Data Entered) ^ ■ I wnw pa UNCLASSIFIED SECURITY CLASSIFICATION OF THIS PAGE(TWi«n Dmtm Enfnd) a supposed match is invalid. Region growing techniques are discussed for extend matching areas into regions of constant parallax and for delimiting uniform regions in an image. Also, two algorithms are presented to show some of the ways in which these techniques can be combined to perform useful tasks in the processing of stereo images.