Automatic and robust single-camera specular highlight removal in cardiac images

Area of Science:

• Medical Imaging
• Computer-Assisted Surgery
• Image Processing

Background:

• Accurate tracking of heart motion is vital in computer-assisted beating heart surgeries.
• Specular reflections on the heart's surface introduce significant errors in motion tracking.

Purpose of the Study:

• To develop a robust method for detecting and removing specular highlights from the heart surface.
• To enhance the accuracy and reliability of heart motion tracking in surgical procedures.

Main Methods:

• A hybrid approach combining color attributes and wavelet-based edge projection for specular highlight detection.
• Dynamic search-based inpainting with adaptive windowing for region recovery.

Main Results:

• The proposed method accurately identifies and removes specular highlights.
• Experimental results confirm the precision and efficiency of the technique.
• The method achieves real-time performance.

Conclusions:

• The developed technique effectively addresses specular reflections, a key challenge in surgical imaging.
• This method offers a generalized solution applicable to various image processing tasks.
• Improved heart motion tracking enhances the safety and efficacy of beating heart surgeries.