Wrong Matches Research Articles

Automated Fast Initial Guess in Digital Image Correlation

ABSTRACTA challenging task that has hampered the fully automatic processing of the digital image correlation (DIC) technique is the initial guess when large deformation and rotation are present. In this paper, a robust scheme combining the concepts of a scale‐invariant feature transform (SIFT) algorithm and an improved random sample consensus (iRANSAC) algorithm is employed to conduct an automated fast initial guess for the DIC technique. The scale‐invariant feature transform algorithm can detect a certain number of matching points from two images even though the corresponding deformation and rotation are large or the images have periodic and identical patterns. After removing the wrong matches with the improved random sample consensus algorithm, the three pairs of closest and non‐collinear matching points serve for the purpose of initial guess calculation. The validity of the technique is demonstrated by both computer simulation and real experiment.

Medical Image Registration Combined with SURF and Improved RANSAC Algorithm

Aiming at the requirements of medical image registration for good robustness, high-accuracy and speed, this paper proposes a Medical Image Registration algorithm Combined SURF with improved RANSAC algorithm. This algorithm first of all extracts SURF featured points from images and matches similar featured points, then the improved RANSAC algorithm is used to eliminate wrong matches, and finally the image registration process is accomplished by estimating space geometric varied parameters according to least square method. The algorithm combines robustness and high efficiency of SURF and high-accuracy of improved RANSAC algorithm. Experimental results show that in the condition of images with noise, non-uniform intensity and large scope of the initial misalignment, the proposed algorithm achieves better robustness and higher speed while maintaining good registration accuracy compared with the conventional ional area-based and feature-based registration methods.

An Image Registration Algorithm Based on Improved SIFT Feature

Aiming at there are long matching time and many wrong matching in the traditional SIFT algorithm, An image registration algorithm based on improved SIFT feature is put forward. First of all, through setting the number of extreme points in the feature point detection, feature points is found according to the DOG space structure from coarse to fine, and the improved SIFT feature descriptor generation algorithm is used. The preliminary matched point pairs are obtained by the nearest neighbor matching criterion, and the bilateral matching method is used for screening the preliminary matched point. Then, the second matching will be done by the similar measurement method based on mahalanobis distance, and RANSAC algorithm is used to calculate the affine transform model. Finally, the transformed image is resampled and interpolated through the bilinear interpolation method. Experimental results show that the algorithm can realize image registration effectively. Image registration technique is an important research content in computer vision and image processing in the, which are widely used in vehicle matching navigation and positioning, cruise missile terminal guidance, target tracking and recognition, image mosaic[1-6]. SIFT algorithm[3-5]can achieve image registration when there are translation, rotation, affine transformation between two images, even for images took by arbitrary angles. And SIFT feature is the milestone of local feature study. But there are long matching time and many wrong matching in the traditional SIFT algorithm, it is difficult to meet the requirement of fast image registration. This paper puts forward an image registration algorithm based on improved SIFT feature, which is robust for image rotation, affine and scale change, and is better than traditional SIFT algorithm.

Research on Wrong Match Pairs Elimination of SIFT Algorithm

The image matching is key technology of image processing applied to many fields. Image matching based on invariant features was the hot spot in image matching research recently. SIFT is one of the most effective scale, rotation and illumination invariant features. However there are a lot of wrong match pairs produced by original SIFT algorithm. Two specific types of wrong matches are analyzed and corresponding eliminating methods are given. Aiming at general wrong matches eliminating, a method based on similar triangles is proposed. Experiments are carried out, and the results confirm that: compared with the art of state, the proposed method is faster; it can eliminate wrong matches cleanly and preserve correct matches as well.

Rapid and accurate face depth estimation in passive stereo systems

In this paper, we introduce a novel approach for face depth estimation in a passive stereo vision system. Our approach is based on rapid generation of facial disparity maps, requiring neither expensive devices nor generic face models. It consists in incorporating face properties into the disparity estimation process to enhance the 3D face reconstruction. We propose a model-based method that is independent from the specific stereo algorithm used. Our method is a two-step process. First, an algorithm based on the Active Shape Model (ASM) is proposed to acquire a disparity model specific to the face concerned. Second, using this model as a guidance, the dense disparity is calculated and the depth map is estimated. Besides, an original post-processing algorithm is proposed in order to detect holes and spikes in the generated depth maps caused by wrong matches and uncertainties. It is based on the smoothness property of the face and a local and global analysis of the image. Experimental results are presented to demonstrate the reconstruction accuracy and the speed of the proposed method.

Kinetic versus Energetic Discrimination in Biological Copying

We study stochastic copying schemes in which discrimination between a right and a wrong match is achieved via different kinetic barriers or different binding energies of the two matches. We demonstrate that, in single-step reactions, the two discrimination mechanisms are strictly alternative and cannot be mixed to further reduce the error fraction. Close to the lowest error limit, kinetic discrimination results in a diverging copying velocity and dissipation per copied bit. On the other hand, energetic discrimination reaches its lowest error limit in an adiabatic regime where dissipation and velocity vanish. By analyzing experimentally measured kinetic rates of two DNA polymerases, T7 and Polγ, we argue that one of them operates in the kinetic and the other in the energetic regime. Finally, we show how the two mechanisms can be combined in copying schemes implementing error correction through a proofreading pathway.

An Improved Map Matching Algorithm

Mistakes often occurs in map matching when road condition is complex, especially there is island ring, or exit and entrance of highway, in vehicle navigation system. Wrong map matching will lead to wrong map display and wrong route guidance, thus system gives a bad impact to driver. In order to solve this problem, this paper includes an improved map matching algorithm based on NDS, which takes curvature and slope into account. This Algorithm coding has been realized by c and proved its availability at last.

Reidentification by Relative Distance Comparison

Matching people across nonoverlapping camera views at different locations and different times, known as person reidentification, is both a hard and important problem for associating behavior of people observed in a large distributed space over a prolonged period of time. Person reidentification is fundamentally challenging because of the large visual appearance changes caused by variations in view angle, lighting, background clutter, and occlusion. To address these challenges, most previous approaches aim to model and extract distinctive and reliable visual features. However, seeking an optimal and robust similarity measure that quantifies a wide range of features against realistic viewing conditions from a distance is still an open and unsolved problem for person reidentification. In this paper, we formulate person reidentification as a relative distance comparison (RDC) learning problem in order to learn the optimal similarity measure between a pair of person images. This approach avoids treating all features indiscriminately and does not assume the existence of some universally distinctive and reliable features. To that end, a novel relative distance comparison model is introduced. The model is formulated to maximize the likelihood of a pair of true matches having a relatively smaller distance than that of a wrong match pair in a soft discriminant manner. Moreover, in order to maintain the tractability of the model in large scale learning, we further develop an ensemble RDC model. Extensive experiments on three publicly available benchmarking datasets are carried out to demonstrate the clear superiority of the proposed RDC models over related popular person reidentification techniques. The results also show that the new RDC models are more robust against visual appearance changes and less susceptible to model overfitting compared to other related existing models.

An Image Matching Method Based on SIFT Feature

Aiming at problems about repeat matching and wrong matching appeared when traditional SIFT algorithm was used in image matching, an image matching method based on SIFT feature was put forward. Firstly, SIFT features were extracted by traditional SIFT algorithm and candidate matching point pairs were obtained by the nearest neighbor rule. Secondly, lateral matching method was used to remove repeat matched dot-pairs. Finally, Mahalanobis distance as a similarity measurement was used to remove wrong matched dot-pairs. Experiment shows this method can achieve image matching effectively with high accuracy.

Triagem da Afinação Vocal: comparação do desempenho de musicistas e não musicistas

To develop a simple and quick-to-apply procedure for pitch-matching scanning that contains vocal imitation tasks of musical sounds of different tones and the temporal ordination of three different tones; to verify its applicability, by comparing the performance of musicians and non-musicians. Participants were 32 adults of both genders without vocal, hearing and/or auditory processing complaints. They were equally divided into two groups: musicians group--MG and non-musicians group--NMG. All participants underwent the Pitch-matching Scanning that included musical stimuli compatible with men and women singing vocal range gathered into two types of tasks: isolate tones and three-tone sequences. Participants were instructed to listen to them and to reproduce them vocally. Voice samples were recorded, analyzed acoustically, and right and wrong matches for the two tasks were characterized. The variables regarding the comparison between groups and types of tasks were statistically analyzed. A difference was found between groups, and the MG presented greater number of right matches in both tasks. The Pitch-matching Scanning was developed and was sensitive to evaluate and compare the performance between groups. Thus, it can be used as a tool for pitch-matching tracking. Musicians presented better performance than non-musicians in the Pitch-matching Scanning.

Automatic Matching Based on Dynamic Threshold in Close-Range Photogrammetry

Points matching is a key technology in close-range photogrammetry. The epipolar line constraint that confines the research area from two dimensions to one dimension, can improve the efficiency of matching process. In the paper, coded targets are applied to get a set of accurate correspondences of points between images, and then the correspondence set is used for calculating fundamental matrices by normalized 8-points algorithm that can be implemented quickly and simply. Because of the ambiguities, a completely accurate correspondence set will not be gained only by epipolar line constraint, and other technologies should be combined with epipolar line constraint. Three views are used to decrease ambiguities, and a method of dynamic threshold is presented to reject wrong correspondences. A proper threshold is calculated, and those correspondences whose corresponding distance is larger than the dynamic threshold, are considered as wrong matching points. Many experiments show that the correct rate of matching is about 99.98%, which means an automatic, efficient, accurate matching process can be implemented by the methods in this paper.

Meaningful Matches in Stereovision

This paper introduces a statistical method to decide whether two blocks in a pair of images match reliably. The method ensures that the selected block matches are unlikely to have occurred "just by chance." The new approach is based on the definition of a simple but faithful statistical background model for image blocks learned from the image itself. A theorem guarantees that under this model, not more than a fixed number of wrong matches occurs (on average) for the whole image. This fixed number (the number of false alarms) is the only method parameter. Furthermore, the number of false alarms associated with each match measures its reliability. This a contrario block-matching method, however, cannot rule out false matches due to the presence of periodic objects in the images. But it is successfully complemented by a parameterless self-similarity threshold. Experimental evidence shows that the proposed method also detects occlusions and incoherent motions due to vehicles and pedestrians in nonsimultaneous stereo.

Vision-Based Aerial Refueling Docking Based on SURF Algorithm

For the Boom and Receptacle Air Refueling, in order to locate the spatial position of the refueling receptacle, an object locating method is developed based on Speeded-up Robust Feature (SURF) algorithm. Firstly, SURF feature vector matching algorithm is used to detect and collect suitable SURF feature points in left and right images produced by binocular stereo vision system separately. Then the point that shows the same spatial position in both left and right images can be located through process such as deleting wrong matching points and calculating the image coordinate of the target point. Finally, the three-dimensional coordinates of the target points could be rebuilt in the camera’s coordinate system. According to the results of experiments, this method has good robustness and practicability.

An Object Tracking Method Based on the Coordinate-Locating of SIFT Algorithm

The SIFT (scale invariant feature transform) algorithm has been successfully used in the image matching field. In this paper, a simplified SIFT algorithm is designed. The number of layers in the Gaussian pyramid is reduced. When it is comparing the keypoints, it uses an outspreading method. The new method can reduce the comparison time and matching time. Although the new algorithm (C-SIFT algorithm) has less matching accuracy than the SIFT algorithm, it adopts a distortion detection method to abandon the wrong matching. Then it uses the coordinate displacement to determine the tracking position. Experimental results show that C-SIFT algorithm can perform steadily and timely.

구름을 포함한 푸쉬브룸 스캐너 영상의 밴드간 상호등록

푸쉬브룸 스캐너 PAN영상과 MS영상 사이에는 오프셋이 존재하며 서로 다른 시간과 각도로 촬영하고 있다. 이로 인하여 구름과 같이 빠르게 움직이는 물체는 오정합 점들을 생성하며 이는 PAN영상과 MS영상간의 상호영상등록의 오차를 발생시킨다. 특히 구름(안개 및 스모그 포함)이 있는 기상조건 하에서 얻어진 위성영상은 구름에 의해 가려진 지형정보를 추출하는 데 있어 많은 문제를 야기하기 때문에 정확한 영상등록을 위해서는 효과적인 구름 탐지 및 제거 알고리즘이 필요하다. 구름 제거를 위한 관련 연구들은 크게 다음과 같은 세 가지로 나누어지는데 (1) 구름 검출 알고리즘을 통해 구름으로 여겨지는 영역을 분리하여 구름영역을 제거하는 방법 (2) 다중분광영상의 밴드정보를 이용하는 방법 (3) 다시기 영상정보를 이용하는 방법들로 나눌 수 있다. 본 연구에서는 구름 지역을 제거하는 방법과 다시기영상을 이용하는 방법을 사용하여 구름이 포함된 푸쉬브룸 스캐너 밴드간 영상등록의 정확도를 비교, 분석하였다. Since PAN(panchromatic) and MS(multispectral) imagery of pushbroom scanner have the offset between PAN and MS CCD(charge coupled device) in the focal plane, PAN and MS images are acquired at different time and angle. Since clouds are fast moving objects, they should lead mis-registration problem with wrong matching points on clouds. The registration of cloudy imagery to recognize and remove the contamination of clouds can be categorized into three classes: (1) cloud is considered as nose and removed (2) employing multi-spectral imagery (3) using multi-temporal imagery. In this paper, method (1) and (3) are implemented and analysed with cloudy pushbroom scanner images.

Computer assisted matching of anatomical vessel trees

Many inexact automatic tree matching algorithms are available. However, they provide matches that are not completely error free. Another option is to use manually matched node-pairs, but this enormously slows down the process. Our contribution to the state of the art is to combine the advantages of both solutions. We enhance the automatic tree matching algorithm designed by Graham et al., so that it is possible to interact with it by previously selecting important matches or by subsequently fixing the provided wrong matches. We apply the tree matching algorithm to the anatomical vasculature of the liver. Furthermore, we developed several visualization features to make manual tree interaction as easy as possible. Both, the interactive and automatic part of the implemented component were evaluated. As a result, the speed of the automatic tree matching algorithm is increased. It takes 7.45s for trees up to 192 nodes and less than 1s if three input matches are provided. In addition to this, an in-depth evaluation of the robustness of the algorithm is presented. The results are remarkable. The average of wrong matches varies between 1.17 and 1.4 node-pairs in the worst cases. The rate of correct matches is high. The evaluation of the visualization features for interactive refinement of matches showed that the percentage of wrong matches found is increased from 56.25% to 78.43%. The mean time to find them is decreased from 227 to 122s.

Novel averaging window f ilter for SIFT in infrared face recognition

The extraction of stable local features directly affects the performance of infrared face recognition algorithms. Recent studies on the application of scale invariant feature transform (SIFT) to infrared face recognition show that star-styled window filter (SWF) can filter out errors incorrectly introduced by SIFT. The current letter proposes an improved filter pattern called Y-styled window filter (YWF) to further eliminate the wrong matches. Compared with SWF, YWF patterns are sparser and do not maintain rotation invariance; thus, they are more suitable to infrared face recognition. Our experimental results demonstrate that a YWF-based averaging window outperforms an SWF-based one in reducing wrong matches, therefore improving the reliability of infrared face recognition systems.

Study on Medical Micro-Image Mosaic with SIFT Features

In order to obtain panoramic view of medical micro-image in remote medical diagnosis system, we should mosaic micro-image sequence accurately in remote node. SIFT features are invariant to image scaling, rotation and translation. The aim of this study is to mosaic micro-image sequence by extracting SIFT features. Firstly, search coordinate of potential features through Gaussian pyramid (with 7 octaves × 6 scales) and then filter pseudo-keypoints. Secondly, describe these features with 128-D vector. The next step is to calculate matching keypoint pairs between two time-successive images according to minimum Euclidean distance. In this step, standard deviation comparing is proposed to eliminate wrong matching pairs and appropriate threshold(8000) through experiment is used to insure matching pairs. Then, construct image motion equation considering rotation and translation and compute motion parameter by solving equation according matching pairs. Finally mosaic all images. These steps are applied in 5 micro-image sequences such as slices of lung tissue, spleen, kidney, frog blood cell and sunflower tissue. The experiment results show that the gap between two images is vanishing and the proposed method can satisfy with medical micro-images sequence mosaic.

Topological localisation based on monocular vision and unsupervised learning

Global localisation is a very fundamental and challenging problem in robotics. This paper presents a new method for mobile robots to recognise scenes with the use of a single camera and natural landmarks. In a learning step, the robot is manually guided on a path. A video sequence is acquired with a font-looking camera. To reduce the perceptual alias of features easily confused, we propose a modified visual feature descriptor which combines colour information and local structure. A location features vocabulary model is built for each individual location by an unsupervised learning algorithm. In the course of travelling, the robot uses each detected interest point to vote for the most likely location. In the case of perceptual aliasing caused by dynamic change or visual similarity, a Bayesian filter is used to increase the robustness of location recognition. Experiments are conducted to prove that application of the proposed feature can largely reduce wrong matches and performance of proposed method is reliable.

Dense stereo matching by region growing based on disparity space sampling

Through the research of the existing region growing method,the authors presented an improved algorithm,which was unlike the existing region growing algorithm that depended on the accuracy of the seeds.The algorithm can continue growing on the smooth region and repetitive patterns region.Firstly,obtain the initial set of seeds in the disparity space by matching the Harris points.Then build the newly growing method in the disparity space.The results demonstrate it can trace out the disparity space.In the growing process,it can rectify the wrong matching automatically,and guarantee matching accuracy and correctness in the smooth and repetitive pattern region.