Photogrammetry: A "low-cost" method to reconstruct the three-dimensional form of small mammals

Abstract

Over the last few years, the field of morphometrics has undergone a profound revolution. During this time, digital imaging technology has made great strides and 3D (three-dimensional) imaging technology has become increasingly available. Nonetheless, many studies of 3D structures still use 2D (two-dimensional) data, even when this may result in the loss of important information. This is particularly so in the study of small mammals, since devices with enough precision for the 3D digitisation of small objects are the most expensive. Thus, the development of low-cost methods aimed at recovering 3D shape from small mammals would be of great interest. Photogrammetry, which allows obtaining 3D data at lower cost than with other 3D techniques, has been widely used in disciplines like geomorphology or architecture. Recently, some studies have started to use photogrammetry to recreate 3D models of animal phenotypes but, to our knowledge, it has not been applied to the study of small mammals or other small vertebrates. In this context, the aim of this study was to test the suitability of photogrammetric techniques to obtain 3D landmarks from mouse skulls as a model system for small mammals. Shape and size of 3D models obtained with photogrammetric techniques were consistent among replicates, even when different sets of photographs were used. The linear measurements obtained from the 3D models were highly correlated with measurements obtained with callipers on actual crania, and differences between both sets of measures were smaller than those among individuals in most of the tested measures. These results show for the first time that photogrammetry is a precise technique for 3D form analysis of small mammals. Photogrammetry also proved to be accurate for obtaining linear measurements between 3D landmarks. However, further studies are needed to demonstrate whether this technique is also accurate at recreating 3D shapes.