Abstract: A method based on the extracting block integrated features of KPCA was presented for the extraction of the ethnic features of face images.Because the complementarity of global features and local features can better reveal the essence of information,firstly,KPCA was used to the extract the global features of face image and the local feature of each sub block,secondly these features was combined into ethnic features,finally the Boosting-RBF classifier designed in the paper was used to identify image samples of ethnic face．The experiment subjects are a constructed Minorities face image database in this paper．The experimental results show that the extracted ethnic features in this paper can identify accurately the ethnic group to which the face image belongs.

Key words: Facial features of ethnic minorities,Feature extraction,BKPCA,Identification of ethnic minorities,RBF