7.2.7 Various Methods For Face Recognition
2.7 Various Methods for Face Recognition: There are various types of face detection algorithms have been developed. Most of the face detection methods take pixel values as features. But they are highly delicate to lighting variations and noises.
Face recognition is the task of identifying an already detected face as a known or unknown face. 2.7.1 PCA:
In this method convert the image training set to image vector. Image vector is used for finding eigen feature weight matrix. Eigen feature can be used for comparing the test samples. It simple to implement. This algorithm is fast and robust. This method speeds up the computational time. This algorithm is failed for different head poses, expressions, and alignments. 2.7.2 SVM:
SVM achieved better detection rate and fewer false alarms. SVM can improve the accuracy and reduce the computation. SVMs treat every (m_n)-pixel image as a point in a mn-dimensional space. Secondly, SVMs compare a candidate point to successive pairs of known classes to determine its experimental class, rather than comparing the distances between the candidate and a series of single points in a high-dimensional space. For a forty-class training set containing ten images of each subject, an SVM facial recognition implementation achieved an average minimum mis classication rate of 3:0% …show more content…
An image is split into smallest regions that LBP histograms are extracted and then concatenated in to a single feature vector. This vector forms an efficient representation of the face area and can be used to measure similarities between images. LBP features are effective and efficient for facial expression recognition. It will take much time. It is easy to implement.
2.7.4
Face recognition is the task of identifying an already detected face as a known or unknown face. 2.7.1 PCA:
In this method convert the image training set to image vector. Image vector is used for finding eigen feature weight matrix. Eigen feature can be used for comparing the test samples. It simple to implement. This algorithm is fast and robust. This method speeds up the computational time. This algorithm is failed for different head poses, expressions, and alignments. 2.7.2 SVM:
SVM achieved better detection rate and fewer false alarms. SVM can improve the accuracy and reduce the computation. SVMs treat every (m_n)-pixel image as a point in a mn-dimensional space. Secondly, SVMs compare a candidate point to successive pairs of known classes to determine its experimental class, rather than comparing the distances between the candidate and a series of single points in a high-dimensional space. For a forty-class training set containing ten images of each subject, an SVM facial recognition implementation achieved an average minimum mis classication rate of 3:0% …show more content…
An image is split into smallest regions that LBP histograms are extracted and then concatenated in to a single feature vector. This vector forms an efficient representation of the face area and can be used to measure similarities between images. LBP features are effective and efficient for facial expression recognition. It will take much time. It is easy to implement.
2.7.4