Scholarlink

<p>A broad spectrum of prevalent conditions associated with imaging and cardiovascular disorders can be effectively identified through advanced image analysis. However, challenges such as low visual saliency, uneven lighting, and obscured images complicate the detection and assessment of low-quality images. To address these challenges and improve visual saliency while assessing image quality in the contexts of compression, transmission, and reconstruction, we introduce a novel enhancement technique termed the linear homotopic parametric value approach. This method has yielded promising outcomes when applied to the structured analysis of publicly accessible image datasets, including MESSIDOR and STARE. Notably, our approach achieved a peak signal-to-noise ratio of 49.9214 and a structured similarity index of 0.9999, outperforming previous studies. These enhancements result in images that are significantly clearer and of higher qualitative value. Our algebraic topology-based enhancement model has been rigorously evaluated across seven distinct quality metrics, consistently achieving superior results and surpassing existing state-of-the-art methodologies, as demonstrated in a comprehensive comparative analysis.</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

Homotopy-based image enhancement techniques for medical imaging applications

  • 期刊论文

  • 发表日期：

    2026年4月2日

查看全文
  • S. Shivam Kumar Jha,
  • N. Mohana

摘要

A broad spectrum of prevalent conditions associated with imaging and cardiovascular disorders can be effectively identified through advanced image analysis. However, challenges such as low visual saliency, uneven lighting, and obscured images complicate the detection and assessment of low-quality images. To address these challenges and improve visual saliency while assessing image quality in the contexts of compression, transmission, and reconstruction, we introduce a novel enhancement technique termed the linear homotopic parametric value approach. This method has yielded promising outcomes when applied to the structured analysis of publicly accessible image datasets, including MESSIDOR and STARE. Notably, our approach achieved a peak signal-to-noise ratio of 49.9214 and a structured similarity index of 0.9999, outperforming previous studies. These enhancements result in images that are significantly clearer and of higher qualitative value. Our algebraic topology-based enhancement model has been rigorously evaluated across seven distinct quality metrics, consistently achieving superior results and surpassing existing state-of-the-art methodologies, as demonstrated in a comprehensive comparative analysis.