Document Image Processing

Abstract

Document Image Processing allows systems like OCR, writer identification, writer recognition, check processing, historical document processing, etc., to extract useful information from document images. What we call a document image ranges from images of historical documents written on various surfaces, to synthetic images (useful for creating datasets) and videos including text. In order to succeed, many preprocessing tasks can be required: document skew detection and correction, slant removal, binarization and segmentation procedures, as well as other normalization tasks. These low-level tasks are generally followed by high-level tasks such as the recognition or spotting of textual elements (characters, words, or text lines). The intent of this Special Issue is to collect the experiences of leading scientists of the field, but also to be an assessment tool for people who are new to the world of document image processing.