| Summary: | 碩士 === 華梵大學 === 電子工程學系碩士班 === 98 === Face recognition has been widely accepted due to its non-intrusive features among biometric identification techniques including fingerprinting and the use of retina, iris, and palm-print imaging. Being the most important preprocessing steps for face recognition in face detection play crucial roles. This algorithm includes certain techniques such as, color segmentation, binary imaging, morphological dilation and erosion operations, as well region growing. First, color segmentation extracts the skin color of a face from a cluttered image; then, binary imaging further forms a more complete region. Next, morphological erosion eliminates some of the small spots in a tested image. Contrary to erosion, dilation enlarges and connects a small and disconnected, but marked, facial region. Subsequently, region growing is employed to mark multiple faces in the image. By the face judgment of symmetry, it could find the angle of center-rotated face. Finally, eyes and nose's geometry relationship are used to validate the corrected facial region. The results of several carefully conducted experiments show that the proposed algorithm works efficiently and accurately for validating multiple facial and oblique facial regions.
Keyword：Skin segmentation, Face detection, Face recognition, Digital image processing, Intelligence video surveillance system .
|
|---|
