A novel approach for extraction of ripple mark parameters based on SfM

Abstract

Ripple marks, formed by non-cohesive materials under the action of water, waves or wind, have been regarded as a diagnostic indicator in sedimentology, especially for the identification of current orientation and paleo-sedimentary environment reconstruction. However, it is difficult to obtain the different parameters of ripple marks in the field. Scientists still hitherto rely on compass and ruler to measure the ripples. In this paper, a method based on Structure from Motion to obtain the parameters of ripples is proposed. The whole workflow can be divided into seven steps: 1) Designing a Structure from Motion survey; 2) Taking photographs of high quality; 3) Generating a low-density point cloud; 4) Creating a high-density point cloud; 5) Building a georeferenced mesh model; 6) Establishing a digital surface model and a digital outcrop model; and 7) Extracting and calculating parameters of ripple marks. Through this approach, 253 attitudes (strike, dip-direction and dip angle) of ripples, 50 ripple wavelengths and 50 ripple heights are extracted. In addition, 50 ripple indexes, 50 ripple symmetry indexes and 50 ripple curvatures are calculated. Compared with traditional field observations, this method could be more efficient, lower-cost, more quantitative and more precise (millimeter scale), and could provide an alternative and competitive manner to extract the surface morphology parameters of consolidated or loose deposits with up to millimeter spatial resolution.