Finding face features in color images using fuzzy hit-or-miss transform
Desislava Dimitrova, Antony T. Popov · 2008
This paper considers the problem of finding facial features in frontal color images. For a given image, our method at first detects skin regions on it using color segmentation in a 2D plane of selected color space (HSV, YES, YCrCb or normalized RGB), and after that generates face candidates. Then, we apply methods of fuzzy mathematical morphology to obtain face candidates in order to find a particular feature - a right eyebrow. A computer implementation in MATLAB is made for experimentation and visual results.