In this paper,a method of FDA for face recognition is presented basing on image segmentation.First,the original images are divided into modular images,which are also called sub-images;then,the well-known FDA method is directly used to the sub-images obtained from the previous step.The advantage of the represented way is that the local discriminant features of the original patterns can be efficiently extracted,and it is really true of the images that have large variations in facial expression and lighting.The ORL face image database is made use of to simulate,and when the training sample is only one,the recognition rate of 90.83 percent is achieved.