Heidelberg Collaboratory for Image Processing, Interdisciplinary Center for Scientific Computing, Heidelberg University, Heidelberg, Germany.
Department of the Languages and Cultures of the Near East, Institute for Assyriology, Heidelberg University, Heidelberg, Germany.
PLoS One. 2020 Dec 16;15(12):e0243039. doi: 10.1371/journal.pone.0243039. eCollection 2020.
The cuneiform script provides a glimpse into our ancient history. However, reading age-old clay tablets is time-consuming and requires years of training. To simplify this process, we propose a deep-learning based sign detector that locates and classifies cuneiform signs in images of clay tablets. Deep learning requires large amounts of training data in the form of bounding boxes around cuneiform signs, which are not readily available and costly to obtain in the case of cuneiform script. To tackle this problem, we make use of existing transliterations, a sign-by-sign representation of the tablet content in Latin script. Since these do not provide sign localization, we propose a weakly supervised approach: We align tablet images with their corresponding transliterations to localize the transliterated signs in the tablet image, before using these localized signs in place of annotations to re-train the sign detector. A better sign detector in turn boosts the quality of the alignments. We combine these steps in an iterative process that enables training a cuneiform sign detector from transliterations only. While our method works weakly supervised, a small number of annotations further boost the performance of the cuneiform sign detector which we evaluate on a large collection of clay tablets from the Neo-Assyrian period. To enable experts to directly apply the sign detector in their study of cuneiform texts, we additionally provide a web application for the analysis of clay tablets with a trained cuneiform sign detector.
楔形文字为我们了解古代历史提供了线索。然而，阅读古老的泥板需要花费大量的时间和多年的训练。为了简化这个过程，我们提出了一种基于深度学习的符号检测器，可以在泥板图像中定位和分类楔形符号。深度学习需要大量的训练数据，这些数据以楔形符号的边界框的形式呈现，但对于楔形文字来说，这些数据并不容易获得，而且获取成本很高。为了解决这个问题，我们利用现有的音译，即拉丁字母表示的泥板内容的符号到符号的表示。由于这些音译没有提供符号定位，因此我们提出了一种弱监督的方法：我们将泥板图像与其对应的音译对齐，以在泥板图像中定位音译符号，然后使用这些本地化的符号代替注释来重新训练符号检测器。一个更好的符号检测器反过来又会提高对齐的质量。我们将这些步骤结合在一个迭代过程中，从而可以仅从音译中训练楔形符号检测器。虽然我们的方法是弱监督的，但少量的注释进一步提高了楔形符号检测器的性能，我们在大量来自新亚述时期的泥板上对其进行了评估。为了使专家能够在对楔形文字文本的研究中直接应用符号检测器，我们还提供了一个带有训练好的楔形符号检测器的泥板分析网络应用程序。
