The system consists of the following components: capturing images using a computer's built-in camera, face detection, storing detected face images in a database, and performing face recognition using input photos. This program implements face detection through a skin color recognition approach, where the facial skin color range is defined as 100≤B≤120 and 140≤R≤160. Pixels within this range are set to white while the remaining pixels are set to black. The algorithm employs the imerode function for spherical erosion and applies median filtering to achieve smoothing effects. Finally, after scaling, binarization, and various processing stages, regions with fewer than 1000 white pixels are discarded. Image segmentation incorporates Euler numbers to eliminate background areas resembling facial colors.
MATLAB
285 views
Tagged