Abstract: This paper proposes the pre-selection sample method of Genetic Algorithm fuzzy C mean. The result is that hold the samples nearing the supper plane, delete the samples far off the supper plane, decrease the training set and the storage. Experiences are based on Residence, and according to reducing samples of original. The decision of reducing number of samples according to SVM's numbers of iterative and SV. The variety of iterative and SV numbers should be little while both segmentation results of reducing and original samples are almost same. This method could be extended to other kind of objects.