Decision Level Fusion of SVM outputs

I have two sets of features predicting the same outputs. But instead of training everything at once by concatenating the features, I would like to train them separately and fuse the decisions. In SVM classification, how can I do that?

Please help me by giving small example or any description.

[Stalin Samuel]

refer http://in.mathworks.com/help/vision/examples/digit-classification-using-hog-features.html