Considering the impact on the selection of matching areas of geomagnetic standard variance, fisher information content and code distortion together, the paper studies a method for selecting suitable matching area by using support vector machine. First, reasonable measure parameters of geomagnetic field were calculated, which could represent geomagnetic information accurately and effectively. Then by coding the measured parameters as input feature vectors of training set, and choosing suitable kernel function and parameters of kernel function to classify the matching areas and non-matching areas. At last, by using testing set, the classified effect was tested. Based on classified precision and the second errors, the parameters of kernel function were picked, and by using geodesic geomagnetic data, the classified results were tested again. The experimental results show that this method holds the capability of flexibility and jamming resistance, and can choice correct geomagnetism aided navigation matching areas.