Robust authentication for paper-based text documents based on text watermarking technology

Abstract

<abstract> <p>Aiming at the problem of easy tampering and difficult integrity authentication of paper text documents, this paper proposes a robust content authentication method for printed documents based on text watermarking scheme resisting print-and-scan attack. Firstly, an authentication watermark signal sequence related to content of text document is generated based on the Logistic chaotic map model; then, the authentication watermark signal sequence is embedded into printed paper document by using a robust text watermarking scheme; finally, the watermark information is extracted from scanned image of paper document, and compared with the authentication watermark information calculated in real time by the text document content obtained by OCR technology, thereby performing content integrity authentication of the paper text documents. Experimental results show that our method can achieve the robust content integrity authentication of paper text documents, and can also accurately locate the tampering position. In addition, the document after embedding the watermark information has a good visual effect, and the text watermarking scheme has a large information capacity.</p> </abstract>

Font Watermarking Network for Text Images

“…In this section, the proposed scheme is compared with existed font-based text watermarking methods Qi et al [19] and Xiao et al [18]. One bit message is embedded in one glyph image in the experiments.…”

“…One bit message is embedded in one glyph image in the experiments. Qi et al [19] embed watermarking messages by shifting the stroke of glyphs which performs better than the usual text watermarking schemes. However, it needs extra manual work to select the specific strokes to generate the codebook and get stuck with arc strokes.…”

“…And its extraction based on image matching leads to the contradiction between the extraction accuracy and visual quality. Thus, we pick the straight strokes to generate the glyph images as the codebook that Qi et al [19] needs in advance. Fig.…”

“…Besides, the burden is heavy due to the fact that each letter needs a network. Qi et al [19] also designed a codebook to embed watermarks by shifting the character strokes horizontally or vertically to get multiple glyphs. Then they extracted watermarks by template matching.…”

Tracing Text Provenance via Context-Aware Lexical Substitution

“…However, it is more challenging to embed watermarks with imperceptible perturbations on text due to its inherent discrete nature. Traditional text watermarking schemes embed watermarks by slightly altering the image features like text format (Brassil, Low, and Maxemchuk 1999;Rizzo, Bertini, and Montesi 2016) and fonts (Xiao, Zhang, and Zheng 2018;Qi et al 2019), which are fragile to cross-media transmissions like OCR. Considering this, natural language watermarking (NLW) schemes choose to manipulate the semantics of text, which are inherently robust in the OCR-style transmissions.…”