Abstract: Abstract: Acocording to the Fisher Linear Discriminant (FLD) extracting feature , this paper recognizes expression with Support Vector Machines (SVM) and the multi-class classification problem is solved by the approach of one-against-one. Experiments of human who participates in test has been trained or not are performed on the JAFFE database, and compared to the nearest classifier, the SVM can get better recognition ratio. So, it is feasible to apply SVM to expression recognition.