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.

Review of Wide-Baseline Stereo Image Matching Based on Deep Learning

Fractal dimension, particle shape, and particle breakage analysis for calcareous sand

3D scanning-based morphological characterization of rammed layer interfaces: a case study of the Ming Great Wall in northwestern China

A common problem with matching algorithms, in photogrammetry and computer vision, is the imperfection of finding all correct corresponding points, so-called inliers, and, thus, resulting in incorrect or mismatched points, so-called outliers. Many algorithms, including the well-known randomized random sample consensus (RANSAC)-based matching, have been developed focusing on the reduction of outliers. RANSAC-based methods, however, have limitations such as increased false positive rates of outliers, and, consequently resulting in fewer inliers, an unnecessary high number of iterations, and high computational time. Such deficiencies possibly result from the random sampling process, the presence of noise, and incorrect assumptions of the initial values. This paper proposes a modified version of RANSAC-based methods, called Empowered Locally Iterative SAmple Consensus (ELISAC). ELISAC improves RANSAC by utilizing three basic modifications individually or in combination. These three modifications are (a) to increase the stability and number of inliers using two Locally Iterative Least Squares (LILS) loops (Basic LILS and Aggregated-LILS), based on the new inliers in each loop, (b) to improve the convergence rate and consequently reduce the number of iterations using a similarity termination criterion, and (c) to remove any possible outliers at the end of the processing loop and increase the reliability of results using a post-processing procedure. In order to validate our proposed method, a comprehensive experimental analysis has been done on two datasets. The first dataset contains the commonly-used computer vision image pairs on which the state-of-the-art RANSAC-based methods have been evaluated. The second dataset image pairs were captured by a drone over a forested area with various rotations, scales, and baselines (from short to wide). The results show that ELISAC finds more inliers with a faster speed (lower computational time) and lower error (outlier) rates compared to M-estimator SAmple Consensus (MSAC). This makes ELISAC an effective approach for image matching and, consequently, for 3D information extraction of very high and super high-resolution imagery acquired by space-borne, airborne, or UAV sensors. In particular, for applications such as forest 3D modeling and tree height estimations where standard matching algorithms are problematic due to spectral and textural similarity of objects (e.g., trees) on image pairs, ELISAC can significantly outperform the standard matching algorithms.

/pdf/an-improved-ransac-outlier-rejection-method-for-uav-derived-3tzc2do0.pdf

An Improved RANSAC Outlier Rejection Method for UAV-Derived Point Cloud

The joint roughness coefficient (JRC), introduced in Barton (1973) represented a new method in rock mechanics and rock engineering to deal with problems related to joint roughness and shear strength estimation. It has the advantages of its simple form, easy estimation, and explicit consideration of scale effects, which make it the most widely accepted parameter for roughness quantification since it was proposed. As a result, JRC has attracted the attention of many scholars who have developed JRC-related methods in many areas, such as geological engineering, multidisciplinary geosciences, mining mineral processing, civil engineering, environmental engineering, and water resources. Because of such a developing trend, an overview of JRC is presented here to provide a clear perspective on the concepts, methods, applications, and trends related to its extensions. This review mainly introduces the origin and connotation of JRC, JRC-related roughness measurement, JRC estimation methods, JRC-based roughness characteristics investigation, JRC-based rock joint property description, JRC's influence on rock mass properties, and JRC-based rock engineering applications. Moreover, the representativeness of the joint samples and the determination of the sampling interval for rock joint roughness measurements are discussed. In the future, the existing JRC-related methods will likely be further improved and extended in rock engineering. 

Advances in joint roughness coefficient (JRC) and its engineering applications

The fractal geometry method has been employed to quantitatively characterize the roughness of a rock discontinuity, which is one of the key factors affecting its shear strength and the seepage characteristics of a rock mass. However, the current fractal methods involving the three-dimensional discontinuity morphology suffer from one or more problems, such as a complicated calculation procedure, an inaccurate calculation result and an inability to characterize the undulation and anisotropy of a discontinuity. To cope with these problems, the discontinuities in artificial granite samples with irregular and undulating surfaces were taken as examples, and a quantitative study on the discontinuity morphology was conducted based on the method of three-dimensional laser scanning in combination with ArcGIS data processing, geographical research, theoretical calculations and regression analysis. After performing systematic research, we proposed an extensive 3D fractal dimension including three discontinuity morphological parameters, i.e. the fractal dimension of discontinuity morphology, the ratio between the maximal undulating amplitude and the discontinuity length, and the average value of all the apparent dip angles of the discontinuity surfaces dipping opposite the shear direction. The extensive 3D fractal dimension can comprehensively characterize the roughness, undulation and anisotropy of the discontinuity morphology. A set of theoretical calculation methods were then developed to determine the three discontinuity morphological parameters of the extensive 3D fractal dimension based on ArcGIS. We finally established a mathematical expression of the extensive 3D fractal dimension. Compared with the current fractal methods, the extensive 3D fractal dimension can effectively compensate for the inability to characterize the undulation and anisotropy of the discontinuity morphology. Its calculation methods have the advantages of simplification, low-time consumption and high precision.

Characterization of discontinuity surface morphology based on 3D fractal dimension by integrating laser scanning with ArcGIS

The invention discloses a shearing box for testing the shearing characteristics of rock mass and soil mass. The shearing box comprises a lower shearing box and an upper shearing box, and the lower shearing box comprises a lower half accommodating cavity with an upper opening; the upper shearing box comprises an upper shearing inner box and an upper shearing outer box, and the upper shearing innerbox is sleeved with the upper shearing outer box; the upper shearing box further comprises an upper connecting plate, when a rock mass shearing test is carried out, an upper transverse plate is in a fixed connection state, the upper connecting plate is fixed to a first position of a right side of the upper shearing outer box, and the upper connecting plate and the upper shearing inner box are in aconnection state; when a soil mass shearing test is carried out, the connecting state of the upper transverse plate is released, the upper connecting plate is fixed to a second position of the rightside of the upper shearing outer box, and the connection between the upper connecting plate and the upper shearing inner box is released. The shear box device is utilized, the requirements of the rockmass shearing test and the soil mass shearing test can be met only through simple modification, and therefore, the test function of indoor rock mass shear equipment or soil mass shear equipment is greatly expanded.

Shearing box is used for testing shearing characteristics of rock mass and soil mass

A shear box for testing shear characteristics of rock mass and soil mass (including hydrate and sediment) is disclosed in the present application. The shear box 5 comprises a lower shear box and an upper shear box. The lower shear box comprises a lower half accommodating cavity with an upper opening; the upper shear box comprises an upper inner shear box and an upper outer shear box, and the upper inner shear box is disposed inside the upper outer shear box. The upper shear box further comprises an upper connecting plate. When a rock mass shear test is performed, an 10 upper horizontal plate is fixedly connected, the upper connecting plate is fixed at a right-side first position of the upper outer shear box, and the upper connecting plate is connected with the upper inner shear box. When a soil mass shear test is performed, the upper horizontal plate is disconnected, the upper connecting plate is fixed at a right-side second position of the upper outer shear box, and the upper connecting plate 15 is disconnected with the upper inner shear box. The present application cause one shear box device to meet the requirements of both rock mass shear tests and soil mass shear tests by making simple changes, thereby greatly expanding the test functions of indoor rock mass shear test equipment or indoor soil mass shear test equipment.

Liu Fangcui

Papers

Characterization of discontinuity surface morphology based on 3D fractal dimension by integrating laser scanning with ArcGIS

Shearing box is used for testing shearing characteristics of rock mass and soil mass

Shear box for testing shear characteristics of rock mass and soil mass