The Recolnat infrastructure supports several programs that allow it to develop the corpus of collections at national and European level, as well as tools and services available to the network.
The e-ReColNat program (ANR-11-INBS-0004) won the 2013 investment program for the future (PIA2) and received 16 million euros over 10 years.
Coordinated by the National Museum of Natural History in partnership with the universities of Montpellier, Burgundy and Clermont Auvergne, IRD, INRAE, CNAM, the Tela Botanica association and the Agoralogie company (with the support of the CNRS and AllEnvi), it was organized in the form of a consortium which defined its governance and project execution methods. This consortium of partners has joined more than 70 institutions in France (mainland and overseas) as data providers.
The main objectives of e-ReColNat were to:
- eventually present images and data from all French naturalist collections on a single platform,
- promote and promote their accessibility for research and expertise on biodiversity and geodiversity.
This platform is fed by existing databases and by digitization programs set up at the national level.
The national consortium, organized in multidisciplinary working groups (WPs), declined and ensured the 3 main axes of the project:
• development of the digital platform;
• aggregation, production and documentation of the corpus of images and data;
• implementation of digital services and tools.
- A consultation portal and a free access database;
- Digitization campaigns for the collections of 35 partner institutions (40 FTEs recruited). These campaigns made it possible to produce: (i) the digitization of 3,800,000 parts of herbariums from 35 institutions, (ii) the computerization of 327,600 type specimens (name tags) and figured specimens in zoology and paleontology, including 162,400 photographed;
- Meta-inventories and censuses of collections: Provence - Alps - Côte-d'Azur, Massif Central, West of France (botany, paleontology, entomology);
- The continuation of the structuring and the provision of additional support for the animation of existing professional networks: Trans'Tyfipal and the network of herbaria de France;
- The implementation of digitization protocols and the organization of training days;
- The development of quality control tools, annotations and image measurements;
- More than 120 publications having used the corpus and/or the tools of the program in different scientific disciplines (systematics, ecology, bioinformatics, social sciences, etc.);
- The consolidation of the functioning of a research infrastructure in the form of a GIS, and its articulation with the organization at European level.
The effort and the means implemented within the framework of e-ReColNat make it possible to present for the first time around 10% of the naturalist collections kept in France.
The e-COL+ project (ANR-21-ESRE-0053) aims to acquire the means to complete and extend the corpus of data and images of specimens from natural history collections produced as part of the PIA2 e-ReColNat . The mastery of 3D equipment now allows the production of large numbers of images and gives access to new data sets for fields such as comparative anatomy, locomotion, food, biomechanics and biomimicry. The project consists of increasing the available fund of 3D models both in surface and microtomography and to associate it with the development of accelerated post-processing thanks to artificial intelligence. The development of innovative methodological channels based on AI, using machine learning and particularly deep learning for character recognition of surface tomography models is one of the primary objectives for this technology.
The 3D models acquired and collected during the project will be made available through a Web interface to ensure the eventual dissemination of all the naturalist data and to ensure the link with European infrastructures. In addition, the services developed will be used to diversify the indexing of images and even see to accelerate and refine post-processing. In return, this will stimulate the dissemination and analysis of collection specimens currently kept in various French museums and institutions.
- Provide state-of-the-art equipment for 3D scanning;
- Build a national repository for the provision of a very large corpus of models: 5000 new models/year on average over the total duration of the project;
- Develop software tools based on artificial intelligence for the recognition and documentation of digital images of specimens;
- Organize the archiving, storage, provision and preservation of digital models.
Articles
Contact
Récolnat 2023