![]() ![]() Here, we address this limitation by introducing an open-source platform for the automated generation of 3D models of arthropods and small objects from a series of 2D images. However, both systems rely on specialised hardware and commercial software for scanner control, image processing and 3D reconstruction, hampering widespread use. To the best of our knowledge, there exist two photogrammetry devices specifically designed for the creation of 3D models of arthropods ( Nguyen et al., 2014 Ströbel et al., 2018). ![]() Perhaps the most promising method for the creation of 3D models is photogrammetric reconstruction, which retains colour information, and represents an excellent compromise between portability, price and quality ( Mathys, Brecko & Semal, 2013 Brecko & Mathys, 2020). As a consequence, the “gold standard” for digitisation are photorealistic and anatomically accurate 3D models ( Wheeler et al., 2012). Even obtaining simple 1D measurements from 2D images is error-prone, due to parallax errors and intra-observer variability, which is larger for measurements obtained from 2D photographs compared to 3D models ( Ströbel et al., 2018 Qian et al., 2019 Brecko & Mathys, 2020). Photographs are doubtlessly useful, but by definition contain substantially less information than the original specimen, as they are restricted to a single image plane ( Nguyen et al., 2014 Ströbel et al., 2018). However, the vast majority of these efforts have focused on high-throughput capture of 2D images, and the convenient automated inclusion of metadata such as barcodes, labels etc. Such a cybertaxonomy has been predicted to revolutionise collaborative taxonomy, and fundamentally change formal and public taxonomic education ( Zhang, Gao & Caelli, 2010 Wheeler et al., 2012). In recognition of these limitations, significant efforts have been underway to digitise natural history collections ( Beaman & Cellinese, 2012 Blagoderov et al., 2012 Mantle, La Salle & Fisher, 2012 Nguyen et al., 2017, 2014 Hudson et al., 2015 Martins et al., 2015 Galantucci, Pesce & Lavecchia, 2016 Erolin, Jarron & Csetenyi, 2017 Ströbel et al., 2018 Galantucci, Guerra & Lavecchia, 2018 Qian et al., 2019 Brecko & Mathys, 2020). This issue is particularly severe for rare and valuable specimens such as holotypes, which can be difficult to access despite their scientific importance. ![]() However, specimen access typically requires to be either physically present on-site, or for specimens to be posted, so reducing the practical value of the collections. Clearly, the utility of these collections hinges on the accessibility of the specimens. Key institutions such as the Natural History Museum in London, the Smithsonian National Museum of Natural History, or the Australian National Insect Collection house upwards of ten million insect specimens, and grow continuously, so archiving part of this diversity ( Mantle, La Salle & Fisher, 2012). The diversity of arthropods is unparalleled ( Misof et al., 2014). The resulting accessibility of scAnt will (i) drive the development of novel and powerful methods for machine learning-driven behavioural studies, leveraging synthetic data (ii) increase accuracy in comparative morphometric studies as well as extend the available parameter space with area and volume measurements (iii) inspire novel forms of outreach and (iv) aid in the digitisation efforts currently underway in several major natural history collections. As a result of the exclusive reliance on generic hardware components, rapid prototyping and open-source software, scAnt costs only a fraction of available comparable systems. We demonstrate how these 3D models can be rigged to enable realistic digital specimen posing, and introduce a novel simple yet effective method to include semi-realistic representations of approximately planar and transparent structures such as wings. The masked images can then be processed further with a photogrammetry software package of choice, including open-source options such as Meshroom, to create high-quality, textured 3D models. These images are then masked with a novel automatic routine which combines random forest-based edge-detection, adaptive thresholding and connected component labelling. scAnt consists of a scanner and a Graphical User Interface, and enables the automated generation of Extended Depth Of Field images from multiple perspectives. We present scAnt, an open-source platform for the creation of digital 3D models of arthropods and small objects. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |