Scholarlink

Screening of neurological disorders can be carried out using various medical image modalities. Outcome of these modalities yields two categories namely: Multi-Spectral (MS) and Panchromatic (PAN) images. The primary aim of medical image fusion is to combine the complementary information in PAN and MS images to minimize redundancy and enhance the diagnostic accuracy. In this paper, the proposed methodology incorporates preprocessing of PAN and MS images with combination of Morphological Top-hat and Bottom-hat transformation operators for contrast enhancement in the first stage. Further, the proposed fusion algorithm involves sub-band decomposition of the images using Discrete Wavelet Transform (DWT). The image coefficients of Low Frequency (LF) band are combined using Average Fusion rule whereas the coefficients from the High Frequency (HF) band are combined using the Max-Max Fusion rule. The resultant fused coefficients are reconstructed using Inverse Discrete Wavelet Transform (IDWT) to restore the structural and textural information of the input image. The quantitative Image Quality Assessment (IQA) of fused images are carried out using some dedicated fusion metrics.

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

Multi-modal Medical Image Fusion Using Wavelets and Morphological Filters for Diagnosis of Neurological Disorders

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Anchal Singh,
  • Vikrant Bhateja,
  • Brij Bihari Tiwari,
  • Amar Singh,
  • Anjoo Patel,
  • Yudong Zhang,
  • Aime’ Lay-Ekuakille,
  • Zaid Omar

摘要

Screening of neurological disorders can be carried out using various medical image modalities. Outcome of these modalities yields two categories namely: Multi-Spectral (MS) and Panchromatic (PAN) images. The primary aim of medical image fusion is to combine the complementary information in PAN and MS images to minimize redundancy and enhance the diagnostic accuracy. In this paper, the proposed methodology incorporates preprocessing of PAN and MS images with combination of Morphological Top-hat and Bottom-hat transformation operators for contrast enhancement in the first stage. Further, the proposed fusion algorithm involves sub-band decomposition of the images using Discrete Wavelet Transform (DWT). The image coefficients of Low Frequency (LF) band are combined using Average Fusion rule whereas the coefficients from the High Frequency (HF) band are combined using the Max-Max Fusion rule. The resultant fused coefficients are reconstructed using Inverse Discrete Wavelet Transform (IDWT) to restore the structural and textural information of the input image. The quantitative Image Quality Assessment (IQA) of fused images are carried out using some dedicated fusion metrics.