Placido Sub-Pixel Edge Detection Algorithm Based on Enhanced Mexican Hat Wavelet Transform and Improved Zernike...

Area of Science:

• Ophthalmology
• Medical Imaging
• Computer Vision

Background:

• Accurate corneal topographic reconstruction requires high-precision localization of the corneal Placido ring edge.
• Existing edge detection algorithms often lack the necessary precision for detailed corneal analysis.

Purpose of the Study:

• To develop a sub-pixel edge detection algorithm for precise corneal Placido ring edge identification.
• To enhance the accuracy and robustness of edge detection in corneal topographic imaging.

Main Methods:

• Employed a multi-scale and multi-position enhanced Mexican Hat Wavelet Transform for initial edge processing.
• Utilized improved Zernike moments for precise edge point relocation using a 9x9 template.
• Implemented two adaptive edge threshold algorithms for final sub-pixel edge point determination.

Main Results:

• The proposed algorithm achieved an average sub-pixel edge error of 0.094 pixels.
• This represents a significant reduction compared to existing algorithms, which had an average error of 0.286 pixels.
• Demonstrated strong robustness against image noise.

Conclusions:

• The novel algorithm effectively meets high-precision location requirements for corneal Placido ring edge detection.
• Offers superior accuracy and noise resilience for corneal topographic reconstruction.
• Provides a valuable tool for advanced ophthalmic imaging analysis.