Chinese Buddhist Literature Database and Image Text Extraction Algorithm

Abstract

The digitization of Chinese Buddhist literature is the key direction of contemporary Buddhist literature research. Currently, there are three types of Chinese Buddhist literature databases that have been completed: Tibetan scriptures databases, non-Tibetan literature databases, and historical materials databases. Commonly used Buddhist literature database image text extraction algorithms are algorithm based on threshold, algorithm based on edge, algorithm based on area, and algorithm based on deep learning. The current Buddhist literature database still needs improvement in terms of repeated construction, image processing, and text recognition and proofreading.