Due to the exponential expansion of information and communication technology, a huge amount of multimedia content is floating over the internet. Among these data, the medical images are most vital and crucial. So, the copyright protection of medical records is the need of the hour. The proposed method presents image watermarking by taking advantage of the redundant discrete wavelet transform (RDWT) and randomized singular value decomposition (RSVD). Fundus images are taken as the input host images, and the watermark is a barcode generated from the string NAMEAGEDISEASECODESEX. The hash value image is generated from the Aadhar number of the patient and media access control (MAC) address. A hybrid watermark is generated by combining the barcode and hash image to enhance the security of the proposed method. The generated hash values are uploaded on ThingSpeak server for real-time internet of medical things (IoMT) identity verification. On an average, imperceptibility is enhanced by 29.51%.

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

CFRV: Copyright Protection of Fundus Images and Real-Time Identity Verification for Smart Healthcare

  • Priyank Khare,
  • Divyanshu Awasthi,
  • Vinay Kumar Srivastava

摘要

Due to the exponential expansion of information and communication technology, a huge amount of multimedia content is floating over the internet. Among these data, the medical images are most vital and crucial. So, the copyright protection of medical records is the need of the hour. The proposed method presents image watermarking by taking advantage of the redundant discrete wavelet transform (RDWT) and randomized singular value decomposition (RSVD). Fundus images are taken as the input host images, and the watermark is a barcode generated from the string NAMEAGEDISEASECODESEX. The hash value image is generated from the Aadhar number of the patient and media access control (MAC) address. A hybrid watermark is generated by combining the barcode and hash image to enhance the security of the proposed method. The generated hash values are uploaded on ThingSpeak server for real-time internet of medical things (IoMT) identity verification. On an average, imperceptibility is enhanced by 29.51%.