Scholarlink

<p>The purpose of this research is to develop a new design of a wavelet interval type-2 takagi–Sugeno-Kang fuzzy brain-imitated neural network (WIT2TFBINN), which is a combination of the mathematical models of a Takagi–Sugeno-Kang (TSK) fuzzy system based on wavelet interval type-2 function (WIT2) and a wavelet interval type-2 fuzzy brain imitated neural network (FBINN). The proposed WIT2TFBINN is used for synchronization control of a 4D Lorentz chaotic system and has the benefits of wavelet interval type-2 membership function, TSK fuzzy inference system, decision making, and emotional activity. To provide fast training, the proposed method's parameter update laws are derived using the gradient descent method. The proposed WIT2TFBINN synchronization technique is then applied to the transmission of medical images in a secure manner. As a cipher image, a medical image is encrypted into a chaotic trajectory. After transmission, the image can be decrypted using chaotic trajectory synchronization on the received signal. The simulation results show that the proposed neural network and the encryption/decryption method are powerful and effective. The results of the static test analysis (histogram, attack analysis, image with noise, image with cropping) show that the encryption/decryption method provides high security.</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

Secure transmission of medical image using a wavelet interval type-2 TSK fuzzy brain-imitated neural network

  • 期刊论文

  • 发表日期：

    2025年2月25日

查看全文
  • Duc-Hung Pham,
  • Tuan-Tu Huynh,
  • Chih-Min Lin,
  • Van Nam Giap,
  • Van-Phong Vu

摘要

The purpose of this research is to develop a new design of a wavelet interval type-2 takagi–Sugeno-Kang fuzzy brain-imitated neural network (WIT2TFBINN), which is a combination of the mathematical models of a Takagi–Sugeno-Kang (TSK) fuzzy system based on wavelet interval type-2 function (WIT2) and a wavelet interval type-2 fuzzy brain imitated neural network (FBINN). The proposed WIT2TFBINN is used for synchronization control of a 4D Lorentz chaotic system and has the benefits of wavelet interval type-2 membership function, TSK fuzzy inference system, decision making, and emotional activity. To provide fast training, the proposed method's parameter update laws are derived using the gradient descent method. The proposed WIT2TFBINN synchronization technique is then applied to the transmission of medical images in a secure manner. As a cipher image, a medical image is encrypted into a chaotic trajectory. After transmission, the image can be decrypted using chaotic trajectory synchronization on the received signal. The simulation results show that the proposed neural network and the encryption/decryption method are powerful and effective. The results of the static test analysis (histogram, attack analysis, image with noise, image with cropping) show that the encryption/decryption method provides high security.