Digitising sandbox experiments using open-source Structure-from-Motion/photogrammetry/Digital Image Correlation package MicMac

Structure-from-Motion and photogrammetry have been implemented in sandbox models to monitor deformation of their surface. Various software packages are available, either open-source of commercial. Some commercial packages are intuitive press-button equipment; these can be very practical to use, but they can be very expensive and unaffordable for most academic laboratories. Another option is open-source alternatives, such as MicMac. MicMac is a versatile open-source package that includes, among others, Structure-from-Motion, photogrammetry and Digital Image Correlation. It has been for various purposes, including geosciences, architecture and biophysics . MicMac has been successfully applied to monitor surface deformation in sandbox models. However, the use of MicMac is not user-friendly, and the steep learning curve of how to use it, and even to install it, can make it repellent for geoscientists that are less familiar with computers and command-line based programs. To overcome this, we provide in this paper a detailed workflow to install and use MicMac as a very effective tool to monitor and digitise surfaces of deforming sandbox models. We will describe how to (1) calibrate synchronized cameras and improve model surfaces for optimal results, (2) install and run MicMac in various OS, and (3) load and plot maps of the digital data (topography, displacements, divergence, shear strain) using Matlab scripts.