The　Gabor　feature　is　effective　for　facial　image　representation.　However,　the　dimension　of　a　Gabor　feature　vector　is　very　high　so　that　the　computation　and　memory　requirements　are　prohibitively　large.　In　this　paper,　we　propose　a　method　to　determine　the　optimal　position　for　extracting　the　Gabor　feature.　The　sub-sampled　positions　of　the　feature　points　are　determined　by　a　mask　generated　from　a　set　of　training　images　by　means　of　Principal　Component　Analysis　(PCA).　With　the　feature　vector　of　reduced　dimension,　a　subspace　LDA　is　applied　for　face　recognition.　Experimental　results　show　that　the　new　sampling　method　is　simple,　and　effective　for　both　dimension　reduction　and　image　representation.　The　recognition　rate　based　on　our　proposed　scheme　is　also　higher　than　that　achieved　using　a　regular　sampling　method　in　a　face　region.　©　2003　IEEE.