Scholarlink

For several diseases, the most reliable modality for identification will be the iris image, especially the diabetes detection. Some publications proved that by examining the iris texture, diabetes can be diagnosed. Considering that, ICNN-based diabetes detection is proposed in this work using iris images. The model includes pre-processing, feature extraction, and detection stages. Firstly, bilateral filtering is used in the pre-processing stage to get the smoothed and noise-removed images. Subsequently, shape features along with LGXP and statistical features are extracted in the second stage. Finally, using these extracted features, the detection process is done by ICNN. The experimental outcomes proved the effectiveness of the ICNN-based diabetes detection approach.

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

Improved Convolutional Neural Network for Diabetes Detection Using Iris Image

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • M. Sandhiya,
  • A. S. Aneetha

摘要

For several diseases, the most reliable modality for identification will be the iris image, especially the diabetes detection. Some publications proved that by examining the iris texture, diabetes can be diagnosed. Considering that, ICNN-based diabetes detection is proposed in this work using iris images. The model includes pre-processing, feature extraction, and detection stages. Firstly, bilateral filtering is used in the pre-processing stage to get the smoothed and noise-removed images. Subsequently, shape features along with LGXP and statistical features are extracted in the second stage. Finally, using these extracted features, the detection process is done by ICNN. The experimental outcomes proved the effectiveness of the ICNN-based diabetes detection approach.