Scholarlink

This research introduces a sophisticated method for the classification and segmentation of human embryo photos, employing a blend of image pre-processing, hybrid clustering methods, and fuzzy adaptive techniques. The dataset, consisting of 840 microscopic pictures categorised into two groups for binary classification, was pre-processed using Adaptive Histogram Equalisation (AHE) and Perona-Malik filtering to improve image quality. Classification methods were utilised before to and after to segmentation, with the classification following segmentation producing more favourable outcomes. Segmentation was executed via the Fuzzy Adaptive Local Information C-Means (FALICM) algorithm, enhanced through multiple meta-heuristic algorithms, such as Particle Swarm Optimization (PSO), Firefly Algorithm (FA), Ant Colony Optimisation (ACO), Grey Wolf Optimiser (GWO), and Bat Algorithm (BA). The BA-FALICM hybrid yielded the most favourable segmentation results. Evaluation metrics including the Dice Similarity Coefficient (DSC), Jaccard Index (IoU), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index (SSIM) exhibited notable enhancements in the accuracy and robustness of embryo image classification and segmentation, highlighting the potential for improved diagnostic precision in medical applications.

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

Enhanced Classification and Segmentation of Human Embryo Images Using Hybrid Fuzzy Clustering and Meta-Heuristic Optimization Techniques

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Vijaya Shree Dharanipathy,
  • B. L. Shivakumar

摘要

This research introduces a sophisticated method for the classification and segmentation of human embryo photos, employing a blend of image pre-processing, hybrid clustering methods, and fuzzy adaptive techniques. The dataset, consisting of 840 microscopic pictures categorised into two groups for binary classification, was pre-processed using Adaptive Histogram Equalisation (AHE) and Perona-Malik filtering to improve image quality. Classification methods were utilised before to and after to segmentation, with the classification following segmentation producing more favourable outcomes. Segmentation was executed via the Fuzzy Adaptive Local Information C-Means (FALICM) algorithm, enhanced through multiple meta-heuristic algorithms, such as Particle Swarm Optimization (PSO), Firefly Algorithm (FA), Ant Colony Optimisation (ACO), Grey Wolf Optimiser (GWO), and Bat Algorithm (BA). The BA-FALICM hybrid yielded the most favourable segmentation results. Evaluation metrics including the Dice Similarity Coefficient (DSC), Jaccard Index (IoU), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index (SSIM) exhibited notable enhancements in the accuracy and robustness of embryo image classification and segmentation, highlighting the potential for improved diagnostic precision in medical applications.