RUTE · Revealing Unseen Text with THz Waves
Horizon 2020 — Marie Skłodowska-Curie Actions
- Duration
- 2022-01-16 → 2024-06-15
- EU contribution
- €183,473
- Participants
- 1
- Scheme
- MSCA-IF
Lines connect the coordinator with its partners.
Results in brief
Revealing Unseen Text with THz Waves
Ancient codices are often fragile, brittle, or damaged, making them difficult to handle without risking permanent damage. In some cases, multiple pages are stacked, rolled, or folded, and the manuscript is too delicate to open. The RUTE project aims to read these "closed books" without opening them, using Terahertz (THz) Time-Domain Spectroscopy and Imaging (THz-TDSI). This non-invasive technique leverages THz radiation's ability to penetrate soft materials like paper and detect unique spectral responses of ancient inks. The main objectives of RUTE are: • O1: Create a unique dataset of THz images of manuscripts using ancient recipes for inks, supports, and bindings. • O2: Develop a pipeline for separating, enhancing, and restoring single-page text from closed books. • O3: Design methods for deblurring and denoising THz images using learned class-specific priors. • O4: Develop an open-source and user-friendly interface for THz image reconstruction. RUTE's methodology, involving beam shape effects removal and the use of text-tailored prior knowledge, aims to significantly enhance the readability of revealed text. The project outcomes will have a profound impact on the analysis, preservation, and digitization of valuable European and global cultural heritage assets. By unlocking hidden texts in ancient codices stored in museums, archives, and private collections, RUTE will uncover secrets concealed for centuries.
Data: CORDIS, © European Union
Project objective
Safely stored in archives’ storages across the world, ancient handwritten books (codices) that are too fragile to be opened, page-turned, or unfolded, are not open to consultation nor accessible, their content hidden from both experts and public. The main goal of the proposed project is to uncover their content and Reveal Unseen Text (RUTE) by reading through them without opening them. Current approaches for facilitating this task are mostly based on techniques that expose fragile manuscripts to ionizing radiation (e.g., X-ray computed tomography) potentially causing damage to ancient inks or support. To prevent damage of valuable manuscripts or written text, RUTE will exploit a non-ionizing technique, terahertz (THz) time-domain imaging, that can penetrate highly absorbing materials (e.g., paper, papyrus, or parchment) and detect paper-air gaps in the sample. Detected paper-air gaps lead to retrieving images that represent a single page of closed books often corrupted by severe system-induced blurring and noise effects. Until now, only limited attempts have been made to reduce these degradation effects (i.e., blur and noise) with none of them tailored to THz images that contain text. To fill this gap, RUTE will facilitate a unique approach to data creation and collection and introduce a highly interdisciplinary methodological process for reading through closed books. The proposed methodology will facilitate detection of paper-air gaps and pages separation, text enhancement, and document images restoration. RUTE will have a tremendous impact on analysis, preservation, and digitisation of valuable European and World cultural heritage assets by enabling the possibility to access knowledge that would otherwise remain undiscovered.
Original text from CORDIS.
Participants
- FONDAZIONE ISTITUTO ITALIANO DI TECNOLOGIA · GenovaCoordinatorItaly
Links
Data: CORDIS, © European Union