Dense Stereo Matching Based on Epipolar Constraint and Edge Points Detecting

孔晓东,屈磊,桂国富,梁栋

DOI: https://doi.org/10.3969/j.issn.1000-3428.2004.20.017

2004-01-01

Abstract:This paper realizes a dense stereo matching between two images based on epipolar constraint and edge points detecting. The result of real experiments proves that this matching can yield relatively dense and exact correspond points.

What problem does this paper attempt to address?

Efficient Large Scale Stereo Matching Based on Cross-Scale.

Yue Xia,Zhitao Liu,Hongye Su,Hao Shu

DOI: https://doi.org/10.1109/icarcv57592.2022.10004232

2022-01-01

Abstract:We propose a binocular stereo matching algorithm CS-ELAS. This is a cross-scale ELAS algorithm that improves the accuracy and robustness of parallax in weakly textured regions and edge regions. Our approach focuses on improving the accuracy and number of support point sets. We uniformly sample the stereo images to obtain candidate support point sets and determine robustly matched support point sets based on an adaptive cross skeleton. In this way a richer and more accurate set of support points can be obtained in the weakly textured areas near the edges. In addition, our method uses the parallax and confidence maps of low-resolution images as a priori for high-resolution images and adds high-confidence pixel information to the set of high-resolution support points. In this way, it not only increases the number of support points in weak texture regions, but also narrows the search range of candidate support points and reduces the computational cost.
Dynamic Limited-Search Matching Algorithm Based on Geometric Properties of Stereo Imaging

王凌云,管业鹏,童林夙,顾伟康,刘济林,叶秀清

DOI: https://doi.org/10.3969/j.issn.1004-1699.2003.03.011

2003-01-01

Abstract:A dynamic limited search matching algorithm based on geometric properties of stereo imaging is developed in this paper. The approximate scope of disparity for matching feature points on the measurement object is ascertained according to the shape of measured object and correlated location spatially between the measured object and target of camera calibration. Based on the geometric properties of stereo imaging, the disparity is obtained by dynamic limited search method. After the disparity computed, the searching orientation and scope of disparity in right (left) image can be determined by way of the disparity difference horizontally between the feature point in left (or right) image and the computed matching point, so the next disparity can be confirmed by calculating correlation value of gray in some region. The rest may be deduced by analogy. In order to gained the correct correspondence, the restriction of geometric comparability should be taken into account due to the instability of gray and much difference from viewpoints also. It is proved that the algorithm proposed in the paper is feasible by computer imitated experiment on standard piece with given 3D coordinates.
Dense stereo matching based on edge constraint and variable windows

Ding Yuan,Feiyang Cheng,Hong Zhang

DOI: https://doi.org/10.1109/ROBIO.2011.6181570

2011-01-01

Abstract:We present a novel method on dense stereo matching; both high accuracy results and a handling of occlusions can be achieved with the edge constraint we prove in this paper. Though a lot of efforts had been made to solve the problems such as occlusions and disparity discontinuities, dense stereo matching is still very challenging in the field of stereo vision. Variable window methods seem to be a good solution, but the existing algorithms usually have complex estimation of an optimal window for each pixel. In our algorithm, the edges detected in the stereo image pair are selected and categorized into two classes to form an efficient constraint, namely edge constraint. By using the edge constraint, the optimal windows for each specific match can be determined and pixels within the windows would be matched with propagation. Compared to the existing window-based algorithms, our method not only can handle the problems caused by occlusions in the stereo image pair, but also can achieve the dense disparity map accurately. Experimental results show the good performance of our method.
Efficient Binocular Stereo Matching Based on Sad and Improved Census Transformation

Yun Zhang,Wenxiang Chen,Han Liu,Jinhua Liu,Hui Du

DOI: https://doi.org/10.1109/icmlc48188.2019.8949324

2019-01-01

Abstract:Binocular stereo matching aims to obtain disparities from two very close views. Existing stereo matching methods may cause false matching when there are much image noise and disparity discontinuities. This paper proposes a novel binocular stereo matching algorithm based on SAD and improved Census transformation. We first perform improved Census transformation, and then we get the matching costs by combining SAD and improved Census transformation. Finally we cluster the matching costs and calculate the disparities. To generate better disparities, we further propose the improved bilateral and selective filters to enhance the accuracy of disparities. Experimental results show that our binocular stereo matching can produce more accurate and complete disparities, and it works well in complex scenes with irregular shapes and more objects, thus it has wide applications in stereoscopic image processing.
Dense Stereo Matching Based on PCNN

舒骁,杨晨晖,刘辉

DOI: https://doi.org/10.1109/icise.2009.454

2009-01-01

Abstract:A key problem in stereo matching lies in selecting an appropriate window size. This paper presents a new method based on using small window for first-step matching and Pulse Coupled Neural Network for perfecting disparity maps. Our algorithm not only reflects the predominance that small window achieves sharper counter, but also gains accurate depth of the region with weak texture and reduces patches effectively. The experimental results indicate that this method could build dense disparity maps with high accuracy compared with common ways.
Dense Stereo Matching Algorithm Based on Image Segmentation

Ma Ruihao,Zhu Feng,Wu Qingxiao,Lu Rongrong,Wei Jingyang

DOI: https://doi.org/10.3788/aos201939.0315001

2019-01-01

Acta Optica Sinica

Abstract:A dense stereo matching algorithm is proposed based on image segmentation. This algorithm combines the gray-gradient algorithm and the zero-mean normalized cross-correlation (ZNCC) algorithm to generate matching cost. The SLIC (Simple Liner Iterative Cluster) algorithm is used for image segmentation. A method based disparity map and superpixels is proposed to update the matching cost. At the disparity post-processing stage, the LRC (Left Right Check), hole filling and cross adaptive window weighted median filtering methods arc used to reduce the error matching rate of the disparity map. The performance evaluation experiments on four Middlebury stereo pairs demonstrate that the proposed algorithm achieves an average error matching rate of 4.99% Key words machine vision; stereo matching algorithm; matching cost computation method fusion; cross adaptive window weighted median filtering
Dense Stereo Matching from Separated Views of Wide-Baseline Images

Qian Zhang,King Ngi Ngan

DOI: https://doi.org/10.1007/978-3-642-17688-3_25

2010-01-01

Abstract:In this paper, we present a dense stereo matching algorithm from multiple wide-baseline images with separated views. The algorithm utilizes the coarse-to-fine strategy to propagate the sparse feature matching to dense stereo for image pixels. First, the images are segmented into non-overlapping homogeneous partitions. Then, in the coarse step, the initial disparity map is estimated by assigning the sparse feature; correspondences, where the spatial location of these features is incorporated with the over-segmentation. The initial occlusion status is obtained by cross-checking test. Finally, the stereo maps are refined by the proposed discontinuity-preserving regularization algorithm, which directly coupling the disparity and occlusion labeling. The experimental results implemented on the real date sets of challenging samples, including the wide-baseline image pairs with both identical scale and different scale, demonstrated the good subjective performance of the proposed method.
Feature Based Stereo Matching Using Two-Step Expansion

Liqiang Wang,Zhen Liu,Zhonghua Zhang

DOI: https://doi.org/10.1155/2014/452803

IF: 1.43

2014-01-01

Mathematical Problems in Engineering

Abstract:This paper proposes a novel method for stereomatching which is based on image features to produce a dense disparity map through two different expansion phases. It can find denser point correspondences than those of the existing seed-growing algorithms, and it has a good performance in short and wide baseline situations. This method supposes that all pixel coordinates in each image segment corresponding to a 3D surface separately satisfy projective geometry of 1D in horizontal axis. Firstly, a state-of-the-art method of feature matching is used to obtain sparse support points and an image segmentation-based prior is employed to assist the first region outspread. Secondly, the first-step expansion is to find more feature correspondences in the uniform region via initial support points, which is based on the invariant cross ratio in 1D projective transformation. In order to find enough point correspondences, we use a regular seed-growing algorithm as the second-step expansion and produce a quasi-dense disparity map. Finally, two different methods are used to obtain dense disparity map from quasi-dense pixel correspondences. Experimental results show the effectiveness of our method.
Enhanced point descriptors for dense stereo matching

Haitao Lang,Yongtian Wang,Xin Qi,Weiqing Pan

DOI: https://doi.org/10.1109/IASP.2010.5476124

2010-01-01

Abstract:We propose a novel local feature descriptor named Enhanced Point Descriptor (referred to as EPD) for dense stereo matching applications. The existing local feature descriptors, e.g., SIFT and SURF, can only be used to represent sparse image extreme points which make stereo matching sparsely. We design EPDs to represent common image points. To generate an EPD, we first build image characteristics vectors for neighborhood points around interest point in a specific sampled window. An EPD is a covariance matrix of characteristics vectors for all sampled points. The image characteristics we used to build vectors include HSV color, Gaussian-weighted gradient norms and orientations, which make EPD robust to rotation, perspective and illumination change. Experimental results show that EPD's performance is superior to commonly used correlation windows methods in dense stereo matching.
An accurate stereo matching method based on color segments and edges

Hui Wei,Lingjiang Meng

DOI: https://doi.org/10.1016/j.patcog.2022.108996

IF: 8

2022-08-29

Pattern Recognition

Abstract:Stereo matching algorithms of binocular vision suffer from low accuracy when dealing with natural scenes (such as industrial robot scenes). Biological vision is sensitive to object edges; it divides objects by their edges, and then perceives their distances. Similar to the biological eye mechanism, this study proposes a matching algorithm that combines segment- and edge-matching to obtain the disparity. In segment matching, pixel strings from the same row of the left and right images are divided into pixel segments, whose colors and lengths are used as clues to determine several types of matching pixel segment pairs according to non-crossing mapping. The analysis of the spatial state yields several types of stimulus bars. Disparities can be obtained from the relation between pixel segment pairs and stimulus bars. In edge matching, the DTW (Dynamic Time Warping) algorithm and the gradient are used to determine the initial edge pixel matching results. The remaining edge point disparity is obtained by fitting a fill to the existing edge point disparity. Finally, segment and edge matching results are combined to check and fill and post-processing. This new matching method transforms pixel matching to pixel segment matching and edge matching, which can reduces the time complexity. The algorithm can be implemented in an industrial robot environment for high-precision needle threading guidance, which neither traditional binocular matching nor deep learning matching algorithms can do.

computer science, artificial intelligence,engineering, electrical & electronic
Densely stereo image matching using SURF and TPS

侯文广,吴梓翠,丁明跃

DOI: https://doi.org/10.13245/j.hust.2010.07.024

2010-01-01

Abstract:Speeded up robust features (SURF) has more chance of finding corresponding feature points, which can be used as seed points for densely matching. Thin-plate spline (TPS) transformation can establish a smooth and continuous transformation relationship between corresponding regions which can be used to estimate parallax in matching. Based on the two methods, the paper proposes a novel matching solution. Firstly, a set of reliable seed points was obtained by filtering the corresponding points obtained using SURF with TPS. Then, the corresponding points of each pixel in left image were estimated by the TPS matrix. Thirdly, the dense match can be carried out between two images. Since the seed points and TPS transformation could estimate the coordinates of corresponding point more accurately, so the dense match can performed easily, reliably and quickly. At the same time, the matching region was confined which is the outer protruding polygon surrounded seed points. Finally, the errors corresponding points using TPS and epipolar restriction were deleted. This new method was clarified using some stereo images in human face reconstruction whose similarity is low very much. Using this method, a dense and reliable point cloud can be obtained.
Stereo Matching Based on Two Cameras and One 3D Image Sensor.

Lu Yang,Xiaowei Shao,Ryosuke Shibasaki,Rongben Wang

DOI: https://doi.org/10.1117/12.872491

2011-01-01

Abstract:Due to the problems of noise, textureless region and depth discontinuity in stereo matching, a new matching method based on two cameras and one 3D image sensor is proposed in this paper. Though the 3D image sensor can offer a depth map, it has low resolution and much noise. Therefore, it can not be used to 3D reconstruction. However, combining two cameras with one 3D image sensor and regarding the depth map as an initial sparse disparity map is an advantage method in stereo vision matching. This method can largely improve the matching accuracy and decrease the running time. Finally, a dense disparity map can be obtained. The experiment results indicate that the proposed algorithm performs well and the disparity map has more accuracy comparing with existing methods.
Research of an improved dense matching algorithm based on graph cuts

Hongwei Gao,Liang Chen,Xiaoyang Liu,Yang Yu

DOI: https://doi.org/10.1109/WCICA.2010.5554641

2010-01-01

Intelligent Control and Automation

Abstract:According to the problem of graph cuts matching algorithm that lots of false matches appears in weak texture area, an improved graph cuts matching algorithm for dense matching is proposed in this paper. First, corner point detection and matching are made between stereo image pairs, and false matches are eliminated by relaxation algorithm and LMeds algorithm. Secondly, triangulate partitioning of discrete matching points is executed, the disparity of every triangle is control by its three vertex. Finally, the dense disparity mapping is draw by graph cuts based on energy function. The experimental results show performance of the proposed algorithm is improved and it will take a shorter time to compute an accuracy dense disparity mapping.
Stereo Matching with Global Edge Constraint and Occlusion Handling.

Feiyang Cheng,Hong Zhang,Ding Yuan,Mingui Sun

DOI: https://doi.org/10.1109/dicta.2012.6411671

2012-01-01

Abstract:In this paper, an effective constraint is proposed to leverage the stereo matching in early vision literature. Firstly, some particular edges are extracted to compose the new smooth constraint by categorizing the color edges into different groups. Then the optimal support windows can be established based on the proposed constraint. Finally, the disparity map would be estimated by using match propagation within the optimal support windows. Compared to the traditional window-based algorithms, our method is more effective to deal with the difficulties caused by the occluded regions or disparity discontinuities. We evaluated our algorithm with real stereo pair of images and the experimental results showed the good performance.
Stereo Matching Based on Phase Congruency Corner Detector

Ding Hui,Fu Mengyin

DOI: https://doi.org/10.1117/12.716942

2006-01-01

Abstract:Feature point matching is the most common one among all kinds of stereo matching. However, since feature points are unique, the disparity map through feature matching is also sparse. In this paper, proposed a dense disparity estimation method that combines the reliability of feature-based correspondence methods and a reliable feature operator. The new operator uses the principal moments of the phase congruency information to determine corner information. The resulting corner operator is highly localized and has responses that are invariant to image contrast. This results in reliable feature detection under varying illumination conditions with fixed thresholds. Selecting those feature points that allow left-right correspondence based on phase correlation surrounding each point. And use the sparse correspondences at feature points as a constraint to control the computation of dense disparity via regularized block matching that minimizes matching and disparity smoothness errors. Experimental results show that this method can eliminate many kinds of outliers effectively.
Stereo Matching Method for Raster Binocular Stereo Vision Sensor

Xie Guanghui,Wei Zhenzhong,Liu Zhen,Zhang Guangjun

DOI: https://doi.org/10.3969/j.issn.1007-2276.2010.02.030

2010-01-01

Abstract:Stereo matching is one of the difficulties of raster binocular stereo vision sensor.For this problem,a stereo matching method was presented based on epipolar constraints and searching minimum distance of space points.The method treated raster stereo vision system as two raster structur light sensors.After calibration,the 3D coordinate for centre point of light stripe to the structure light model was measured by using this method.If two points were matching points,their distance became zero.Leading in epipolar constraint,a feature point could be found in the left picture.Then a epipolar line could be found correspondingly in the right picture.The matching point could be found in the intersection points of epipolar line and structure line stripes.This method achieved points matching in 3D space.It has small account of calculation,and can obtain unique matching point.The simulation result shows the effectiveness of the method.
Effective Stereo Matching Using Reliable Points Based Graph Cut

Hao-qian Wang,Mian Wu,Yong-bing Zhang,Lei Zhang

DOI: https://doi.org/10.1109/vcip.2013.6706415

2013-01-01

Abstract:In this paper, we propose an effective stereo matching algorithm using reliable points and region-based graph cut. Firstly, the initial disparity maps are calculated via local window-based method. Secondly, the unreliable points are detected according to the DSI(Disparity Space Image) and the estimated disparity values of each unreliable point are obtained by considering its surrounding points. Then, the scheme of reliable points is introduced in region-based graph cut framework to optimize the initial result. Finally, remaining errors in the disparity results are effectively handled in a multi-step refinement process. Experiment results show that the proposed algorithm achieves a significant reduction in computation cost and guarantee high matching quality.
Dense Stereo Correspondence with Contrast Context Histogram, Segmentation-Based Two-Pass Aggregation and Occlusion Handling

Tianliang Liu,Pinzheng Zhang,Limin Luo

DOI: https://doi.org/10.1007/978-3-540-92957-4_39

2009-01-01

Abstract:In a local and perceptual organization framework, a novel stereo correspondence algorithm is proposed to provide dense and accurate disparity maps under point ambiguity. First, the initial matching technique is based on raw matching cost obtained from local descriptor with contrast context histogram and two-pass cost aggregation via segmentation-based adaptive support weight. Second, the disparity estimation procedure consists sequentially of two steps: namely, a narrow occlusion handling and a multi-directional weighted least square (WLS) fitting for large occlusion. The experiment results indicate that our algorithm can increase robustness against outliers, and then obtain comparable and accurate disparity than other local stereo methods effectively, and it is even better than some algorithms using advanced and offline but computationally complicated global optimization based algorithms.
Stereo matching by using the global edge constraint

Feiyang Cheng,Hong Zhang,Ding Yuan,Mingui Sun

DOI: https://doi.org/10.1016/j.neucom.2013.10.022

IF: 6

2014-01-01

Neurocomputing

Abstract:Stereo matching, the key problem in the field of computer vision has long been researched for decades. However, constructing an accurate dense disparity map is still very challenging for both local and global algorithms, especially when dealing with the occlusions and disparity discontinuities. In this paper, by exploring the characteristics of the color edges, a novel constraint named the global edge constraint (GEC) is proposed to discriminate the locations of potential occlusions and disparity discontinuities. The initial disparity map is estimated by using a local algorithm, in which the GEC could guarantee that the optimal support windows would not cross the occlusions. Then a global optimization framework is adopted to improve the accuracy of the disparity map. The data term of the energy function is constructed by using the reliable correspondences selected from the initial disparity map; and the smooth term incorporates the GEC as a soft constraint to handle the disparity discontinuities. Optimal solution can be approximated via existing energy minimization approaches such as Graph cuts used in this paper. Experimental results using the Middlebury Stereo test bed demonstrate the superior performance of the proposed approach.
Feature Matching Based on Geometric Constraints in Weakly Calibrated Stereo Views of Curved Scenes

Bian Houqin,Su Jianbo

DOI: https://doi.org/10.1016/s1004-4132(08)60121-8

2008-01-01

Abstract:The identification of the correspondences of points of views is an important task. A new feature matching algorithm for weakly calibrated stereo images of curved scenes is proposed, based on mere geometric constraints. After initial correspondences are built via the epipolar constraint, many point-to-point image mappings called homographies are set up to predict the matching position for feature points. To refine the predictions and reject false correspondences, four schemes are proposed. Extensive experiments on simulated data as well as on real images of scenes of variant depths show that the proposed method is effective and robust.

Dense Stereo Matching Based on Epipolar Constraint and Edge Points Detecting

Efficient Large Scale Stereo Matching Based on Cross-Scale.

Dynamic Limited-Search Matching Algorithm Based on Geometric Properties of Stereo Imaging

Dense stereo matching based on edge constraint and variable windows

Efficient Binocular Stereo Matching Based on Sad and Improved Census Transformation

Dense Stereo Matching Based on PCNN

Dense Stereo Matching Algorithm Based on Image Segmentation

Dense Stereo Matching from Separated Views of Wide-Baseline Images

Feature Based Stereo Matching Using Two-Step Expansion

Enhanced point descriptors for dense stereo matching

An accurate stereo matching method based on color segments and edges

Densely stereo image matching using SURF and TPS

Stereo Matching Based on Two Cameras and One 3D Image Sensor.

Research of an improved dense matching algorithm based on graph cuts

Stereo Matching with Global Edge Constraint and Occlusion Handling.

Stereo Matching Based on Phase Congruency Corner Detector

Stereo Matching Method for Raster Binocular Stereo Vision Sensor

Effective Stereo Matching Using Reliable Points Based Graph Cut

Dense Stereo Correspondence with Contrast Context Histogram, Segmentation-Based Two-Pass Aggregation and Occlusion Handling

Stereo matching by using the global edge constraint

Feature Matching Based on Geometric Constraints in Weakly Calibrated Stereo Views of Curved Scenes