Search Results (7)

A binocular stereo vision measurement system is widely used in fields such as industrial inspection and marine engineering due to its high accuracy, low cost, and ease of deployment. An unreasonable structural design can lead to difficulties in image matching and inaccuracies in depth computation during subsequent processing, thereby limiting the system’s performance and applicability. This paper establishes a systemic error analysis model to enable the validation of changes in structural parameters on the performance of the binocular vision measurement. Specifically, the impact of structural parameters such as baseline distance and object distance on measurement error is analyzed. Extensive experiments reveal that when the ratio of baseline length to object distance is between 1 and 1.5, and the angle between the baseline and the optical axis is between 30 and 40 degrees, the system measurement error is minimized. The experimental conclusions provide guidance for subsequent measurement system research and parameter design. Full article

To promote the development of deep learning for feature matching, image registration, and three-dimensional reconstruction, we propose a method of constructing a deep learning benchmark dataset for affine-invariant feature matching. Existing images often have large viewpoint differences and areas with weak texture, which may cause difficulties for image matching, with respect to few matches, uneven distribution, and single matching texture. To solve this problem, we designed an algorithm for the automatic production of a benchmark dataset for affine-invariant feature matching. It combined two complementary algorithms, ASIFT (Affine-SIFT) and LoFTR (Local Feature Transformer), to significantly increase the types of matching patches and the number of matching features and generate quasi-dense matches. Optimized matches with uniform spatial distribution were obtained by the hybrid constraints of the neighborhood distance threshold and maximum information entropy. We applied this algorithm to the automatic construction of a dataset containing 20,000 images: 10,000 ground-based close-range images, 6000 satellite images, and 4000 aerial images. Each image had a resolution of 1024 × 1024 pixels and was composed of 128 pairs of corresponding patches, each with 64 × 64 pixels. Finally, we trained and tested the affine-invariant deep learning model, AffNet, separately on our dataset and the Brown dataset. The experimental results showed that the AffNet trained on our dataset had advantages, with respect to the number of matching points, match correct rate, and matching spatial distribution on stereo images with large viewpoint differences and weak texture. The results verified the effectiveness of the proposed algorithm and the superiority of our dataset. In the future, our dataset will continue to expand, and it is intended to become the most widely used benchmark dataset internationally for the deep learning of wide-baseline image matching. Full article

This paper presents research on 3D scanning by taking advantage of a camera array consisting of up to five adjacent cameras. Such an array makes it possible to make a disparity map with a higher precision than a stereo camera, however it preserves the advantages of a stereo camera such as a possibility to operate in wide range of distances and in highly illuminated areas. In an outdoor environment, the array is a competitive alternative to other 3D imaging equipment such as Structured-light 3D scanners or Light Detection and Ranging (LIDAR). The considered kinds of arrays are called Equal Baseline Camera Array (EBCA). This paper presents a novel approach to calibrating the array based on the use of self-calibration methods. This paper also introduces a testbed which makes it possible to develop new algorithms for obtaining 3D data from images taken by the array. The testbed was released under open-source. Moreover, this paper shows new results of using these arrays with different stereo matching algorithms including an algorithm based on a convolutional neural network and deep learning technology. Full article

Strong geometric and radiometric distortions often exist in optical wide-baseline stereo images, and some local regions can include surface discontinuities and occlusions. Digital photogrammetry and computer vision researchers have focused on automatic matching for such images. Deep convolutional neural networks, which can express high-level features and their correlation, have received increasing attention for the task of wide-baseline image matching, and learning-based methods have the potential to surpass methods based on handcrafted features. Therefore, we focus on the dynamic study of wide-baseline image matching and review the main approaches of learning-based feature detection, description, and end-to-end image matching. Moreover, we summarize the current representative research using stepwise inspection and dissection. We present the results of comprehensive experiments on actual wide-baseline stereo images, which we use to contrast and discuss the advantages and disadvantages of several state-of-the-art deep-learning algorithms. Finally, we conclude with a description of the state-of-the-art methods and forecast developing trends with unresolved challenges, providing a guide for future work. Full article

Among surrounding information-gathering devices, cameras are the most accessible and widely used in autonomous vehicles. In particular, stereo cameras are employed in academic as well as practical applications. In this study, commonly used webcams are mounted on a vehicle in a dual-camera configuration and used to perform lane detection based on image correction. The height, baseline, and angle were considered as variables for optimizing the mounting positions of the cameras. Then, a theoretical equation was proposed for the measurement of the distance to the object, and it was validated via vehicle tests. The optimal height, baseline, and angle of the mounting position of the dual camera configuration were identified to be 40 cm, 30 cm, and 12°, respectively. These values were utilized to compare the performances of vehicles in stationary and driving states on straight and curved roads, as obtained by vehicle tests and theoretical calculations. The comparison revealed the maximum error rates in the stationary and driving states on a straight road to be 3.54% and 5.35%, respectively, and those on a curved road to be 9.13% and 9.40%, respectively. It was determined that the proposed method is reliable because the error rates were less than 10%. Full article

This paper presents an active wide-baseline triple-camera measurement system designed especially for 3D modeling in general outdoor environments, as well as a novel parallel surface refinement algorithm within the multi-view stereo (MVS) framework. Firstly, the pre-processing module converts the synchronized raw triple images from one single-shot acquisition of our setup to aligned RGB-Depth frames, which are then used for camera pose estimation using iterative closest point (ICP) and RANSAC perspective-n-point (PnP) approaches. Afterwards, an efficient dense reconstruction method, mostly implemented on the GPU in a grid manner, takes the raw depth data as input and optimizes the per-pixel depth values based on the multi-view photographic evidence, surface curvature and depth priors. Through a basic fusion scheme, an accurate and complete 3D model can be obtained from these enhanced depth maps. For a comprehensive test, the proposed MVS implementation is evaluated on benchmark and synthetic datasets, and a real-world reconstruction experiment is also conducted using our measurement system in an outdoor scenario. The results demonstrate that (1) our MVS method achieves very competitive performance in terms of modeling accuracy, surface completeness and noise reduction, given an input coarse geometry; and (2) despite some limitations, our triple-camera setup in combination with the proposed reconstruction routine, can be applied to some practical 3D modeling tasks operated in outdoor environments where conventional stereo or depth senors would normally suffer. Full article

Line matching is the foundation of three-dimensional (3D) outline reconstruction for city buildings in aerial photogrammetry. Many existing studies have good line matching effects when dealing with aerial images with short baselines and small viewing angles. However, when faced with wide-baseline and large viewing-angle images, the matching effect drops sharply or even fails altogether. This paper deals with an efficient and simple method to achieve better line matching performance by a pair of wide-baseline aerial images, which make use of viewpoint-in variance to conduct line matching in rectified image spaces. Firstly, the perspective transformation relationship between the image plane and the geoid plane can be established from a Positioning and Orientation System (POS). Then, according to perspective projection matrices, two original images are separately rectified to conformal images, whose perspective deformation of large viewing-angle can be eliminated. Finally, the rectified images are used to conduct line matching, and the matched line segments obtained are back-projected to the original images. Four pairs of urban oblique aerial images are used to demonstrate the validity and efficiency of this method. Compared with line matching on original images, the number and the correctness of the matched line segments are greatly improved. Moreover, there is no loss of time efficiency. The proposed method can also be applied to general UAV (Unmanned Aerial Vehicle) aerial photogrammetry and introduced into matching for other geometric features, such as points, circles, curves, etc. Full article