Scholarlink

<p>Existing approaches that map multiple hash sequences to a single cover element typically use an entire image as the cover. As a result, each image can represent only one hash sequence at a time, leading to the need for transmitting a large number of cover images during the hiding process. Although an image can technically represent multiple hash sequences, its effective capacity is limited to the length <i>L</i> of a single sequence. Conversely, methods that use image blocks as cover elements rely on one-to-one mapping between sequences and blocks, causing the number of required blocks to grow exponentially with the sequence length. Moreover, they lack a secure and well-defined method for transmitting the location information required for message extraction. This paper proposes a novel approach that combines block-level hash encoding with sequence category mapping. By categorizing sequences and mapping a sequence category to a single image block, the method increases the effective capacity of an image, allowing a single image to convey a full message. Additionally, the number of required cover blocks reduces from <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(2^L\)</EquationSource> </InlineEquation> to <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(L+1\)</EquationSource> </InlineEquation>. The secure and practical transmission of location information is ensured using a distributed data hiding technique in a single cloud environment. Our method, dubbed WYSAWIS (What You See And What It iS), highlights how seemingly innocuous image and file transfers may hide data. Compared to other coverless image steganography approaches, WYSAWIS allows grouping up to 6,435 sequences under one cover block, which reduces the number of cover blocks from 32,768 to just 16.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A high-capacity coverless image steganography approach using block-level hash encoding and sequence category mapping: WYSAWIS

  • 期刊论文

  • 发表日期：

    2025年9月17日

查看全文
  • Damaris Belle M. Fotso,
  • Serge Alain Ebele,
  • Stéphane Gael R. Ekodeck,
  • René Ndoundam

摘要

Existing approaches that map multiple hash sequences to a single cover element typically use an entire image as the cover. As a result, each image can represent only one hash sequence at a time, leading to the need for transmitting a large number of cover images during the hiding process. Although an image can technically represent multiple hash sequences, its effective capacity is limited to the length L of a single sequence. Conversely, methods that use image blocks as cover elements rely on one-to-one mapping between sequences and blocks, causing the number of required blocks to grow exponentially with the sequence length. Moreover, they lack a secure and well-defined method for transmitting the location information required for message extraction. This paper proposes a novel approach that combines block-level hash encoding with sequence category mapping. By categorizing sequences and mapping a sequence category to a single image block, the method increases the effective capacity of an image, allowing a single image to convey a full message. Additionally, the number of required cover blocks reduces from \(2^L\) to \(L+1\) . The secure and practical transmission of location information is ensured using a distributed data hiding technique in a single cloud environment. Our method, dubbed WYSAWIS (What You See And What It iS), highlights how seemingly innocuous image and file transfers may hide data. Compared to other coverless image steganography approaches, WYSAWIS allows grouping up to 6,435 sequences under one cover block, which reduces the number of cover blocks from 32,768 to just 16.