Busy for the most part of the semester fingerprint, palm processing, here a little to sort out some
Goal
Find the center point and Triangle Point of fingerprint image, count the number of interval ridge lines, complete the fingerprint classification.
Function
The processing that needs to be done includes:
- Segmentation foreground and background (using gradient map, direction field realization);
- Enhanced fingerprint image (through equalization, Edge convergence, smoothing, enhancement, two-finger);
- Refinement (extraction of single-pixel ridge lines)
- Mark feature points (slit method for direction map, feature points based on Poincare index)
- Statistical connection and Ridge line intersection
Effect
Fingerprint type identification:
Ridge line Number Statistics:
Follow-up supplement (RELATED links)
Some concrete methods. I'll add it here when I'm free.
- Http://www.cnblogs.com/KC-Mei/p/4335939.html of "MATLAB" binary effect
- "MATLAB" Edge extraction effect http://www.cnblogs.com/KC-Mei/p/4335922.html
- "MATLAB" Morphological transformation http://www.cnblogs.com/KC-Mei/p/4335955.html
- "MATLAB" image thinning algorithm http://www.cnblogs.com/KC-Mei/p/4335975.html
"MFC" fingerprint type recognition (OpenCV + matlab mixed programming)