Computerized Paleography

The relationship between the expansion of literacy in Judah and composition of biblical texts has attracted scholarly attention for over a century. Information on this issue can be deduced from Hebrew inscriptions from the final phase of the first Temple period. We report our investigation of 16 inscriptions from the Judahite desert fortress of Arad, dated ca. 600 BCE—the eve of Nebuchadnezzar’s destruction of Jerusalem. The inquiry is based on new methods for image processing and document analysis, as well as machine learning algorithms. These techniques enable identification of the minimal number of authors in a given group of inscriptions. Our algorithmic analysis, complemented by the textual information, reveals a minimum of six authors within the examined inscriptions. The results indicate that in this remote fort literacy had spread throughout the military hierarchy, down to the quartermaster and probably even below that rank. This implies that an educational infrastructure that could support the composition of literary texts in Judah already existed before the destruction of the first Temple. A similar level of literacy in this area is attested again only 400 y later, ca. 200 BCE.

The problem of finding a prototype for typewritten or handwritten characters belongs to a family of "shape prior" estimation problems. In epigraphic research, such priors are derived manually, and constitute the building blocks of "paleographic tables". Suggestions for automatic solutions to the estimation problem are rare in both the Computer Vision and the OCR/Handwriting Text Recognition communities. We review some of the existing approaches, and propose a new robust scheme, suitable for the challenges of degraded historical documents. This fast and easy to implement method is employed for ancient Hebrew inscriptions dated to the First Temple period.

During the 2010 excavation season, two inscriptions showing Hieratic signs written with black ink were found at Qubur el-Walaydah in the coastal plain of southern Israel. Although unearthed recently, new color photographs show that their legibility has deteriorated compared to those taken shortly after excavation. We imaged these inscriptions using multispectral (MS) techniques described by (Faigenbaum and Sober et al. 2012). Notwithstanding the fading of the ink, these MS images significantly improved the inscriptions legibility. In several cases, faded and unrecognizable letters are clearly seen. The new MS photographs, which were essential in deciphering the inscriptions, are presented.

This article surveys an ongoing research of the Legibility Enhancement of Ostraca (LEO) team of Tel Aviv University in the field of computerized paleography of Hebrew Iron Age ink-written ostraca. We perform paleographic tasks using tools from the fields of image processing and machine learning. Several new techniques serving this aim, as well as an adaptation of existing ones, are described herein. This includes testing a range of signal-acquisition methodologies, out of which multispectral imaging and Raman spectroscopy have matured into imaging systems. In addition, we deal with semi- or fully automated facsimile construction and refinement, facsimile, and character evaluation, as well as the reconstruction of broken character strokes. We conclude with future research directions, addressing some of the long-standing epigraphic questions, such as the number of scribes in specific corpora or detection of chronological concurrences and inconsistencies.

Document binary images, created by different algorithms, are
commonly evaluated based on a pre-existing ground truth.
Previous research found several pitfalls in this methodology and
suggested various approaches addressing the issue. This article
proposes an alternative binarization quality evaluation solution for binarized glyphs, circumventing the ground truth. Our method relies on intrinsic properties of binarized glyphs. The features used for quality assessment are stroke width consistency, presence of small connected components (stains), edge noise, and the average edge curvature. Linear and tree-based combinations of these features are also considered. The new methodology is tested and shown to be nearly as sound as human experts’ judgments.

This article surveys an ongoing research of the Legibility Enhancement of Ostraca (LEO) team of Tel Aviv University in the field of computerized paleography of Hebrew Iron Age ink-written ostraca. We perform paleographic tasks using tools from the fields of image processing and machine learning. Several new techniques serving this aim, as well as an adaptation of existing ones, are described herein. This includes testing a range of signal-acquisition methodologies, out of which multispectral imaging and Raman spectroscopy have matured into imaging systems. In addition, we deal with semi- or fully automated facsimile construction and refinement, facsimile, and character evaluation, as well as the reconstruction of broken character strokes. We conclude with future research directions, addressing some of the long-standing epigraphic questions, such as the number of scribes in specific corpora or detection of chronological concurrences and inconsistencies.