As an important image processing technology, face recognition is an active subject in the fields of biometrics, but the low recognition rate hampers its wider use. In order to improve the operation speed and precision, a modified PCA(Principal Component Analysis) based face recognition algorithm is proposed, before applying PCA transform, background removal operation based on skin-color detection is done first, simulation results show that the better efficiency can be achieved by the improved PCA method. Therefore, it is valid in face recognition.
展开▼