研究目的
To improve the visual quality of low-contrast fundus images by proposing two-stage histogram enhancement schemes using cascaded FHBE and CLAHE algorithms.
研究成果
The cascaded FHBE-CLAHE and CLAHE-FHBE schemes, as well as the fusion method, outperform individual enhancement algorithms in improving the visual quality of fundus images, with FHBE-CLAHE showing the best results. Future work should include objective metrics, expert evaluations, and comparisons with other algorithms.
研究不足
The study relies solely on visual quality assessment without objective metrics or expert subjective evaluation. It does not compare with other enhancement algorithms, and the fusion rule may require optimization for further improvement.
1：Experimental Design and Method Selection:
The study employs a cascaded approach combining FHBE and CLAHE algorithms for two-stage enhancement. FHBE is based on fuzzy logic and histogram stretching in HSV color space, while CLAHE uses adaptive histogram equalization with clipping in RGB color space.
2：Sample Selection and Data Sources:
Fundus images are sourced from the Standard Diabetic Retinopathy database (DIARETDB), specifically DIARETDB0 and DIARETDB1, which include normal and retinopathy-affected images.
3：List of Experimental Equipment and Materials:
No specific equipment is mentioned; the methods are algorithmic and implemented in software.
4：Experimental Procedures and Operational Workflow:
Images are converted to HSV for FHBE (enhancing V component) or RGB for CLAHE (enhancing green channel). For two-stage schemes, one algorithm is applied first, and its output is fed to the second. A fusion method is also tested by combining V components of individually enhanced images using a neighborhood-based rule.
5：Data Analysis Methods:
Analysis is based on visual quality assessment; no objective metrics or statistical techniques are specified.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
