Scholarlink

<p>With the increasing use of telemedicine in healthcare, the security and integrity of medical images during transmission have become critical. This paper presents a novel blind watermarking scheme Local Binary Pattern–Discrete Wavelet Transform (LBP–DWT) for medical images based on Local Binary Patterns and the Discrete.Wavelet Transform in frequency domain. We take advantage of the LBP, which is computationally fast, to improve the watermark's resistance to the different kinds of attacks, and maintain the overall visual quality of the watermarked images. In addition, the DWT could offer a high trade-off between robustness and imperceptibility due to the multi-resolution analysis it provides. During embedding, the <i>LL</i> band (approximation coefficients) of the DWT is selected and divided into 3 × 3 blocks. The resulting LBP codes are then XORed with the embedding bits and hidden in the corresponding blocks using the Least Significant Bit technique. Note that the Arnold transform is used during the embedding step to scramble the watermark, which is then vectorized based on the ZigZag fashion to improve the security of the proposed scheme. To evaluate the performance of the proposed method, extensive experiments are conducted on a dataset of medical images. The watermarked images are tested against various attacks, including compression, noise addition, and cropping. The obtained results demonstrate the effectiveness of the proposed techniques.</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

Blind Medical Image Watermarking Based on LBP–DWT for Telemedicine Applications

  • 期刊论文

  • 发表日期：

    2025年3月8日

查看全文
  • Khaled Hebbache,
  • Oussama Aiadi,
  • Belal Khaldi,
  • Ali Benziane

摘要

With the increasing use of telemedicine in healthcare, the security and integrity of medical images during transmission have become critical. This paper presents a novel blind watermarking scheme Local Binary Pattern–Discrete Wavelet Transform (LBP–DWT) for medical images based on Local Binary Patterns and the Discrete.Wavelet Transform in frequency domain. We take advantage of the LBP, which is computationally fast, to improve the watermark's resistance to the different kinds of attacks, and maintain the overall visual quality of the watermarked images. In addition, the DWT could offer a high trade-off between robustness and imperceptibility due to the multi-resolution analysis it provides. During embedding, the LL band (approximation coefficients) of the DWT is selected and divided into 3 × 3 blocks. The resulting LBP codes are then XORed with the embedding bits and hidden in the corresponding blocks using the Least Significant Bit technique. Note that the Arnold transform is used during the embedding step to scramble the watermark, which is then vectorized based on the ZigZag fashion to improve the security of the proposed scheme. To evaluate the performance of the proposed method, extensive experiments are conducted on a dataset of medical images. The watermarked images are tested against various attacks, including compression, noise addition, and cropping. The obtained results demonstrate the effectiveness of the proposed techniques.