Abstract
In this paper, a method for classifying electroencephalographic (EEG) recordings with images as stimulation is introduced, which aims at selecting the target images. EEG recordings to be processed are referred to the onset of the test images with a single stimulation so as to avoid spending extra time on repeating images. Independent component analysis (ICA) is used to reduce the redundancy of EEG recordings, and wavelet packet (WP) analysis is efficient for dealing with the non-stationary character of brain activity. Feature vectors are extracted by a method that combines these two algorithms. The support vector machine is used as a classifier, carrying out the classification result. The experimental results demonstrate that the accuracy of this method's image classification is affected very little by different classifier parameters. The best result achieves 90% accuracy, which indicts it is feasible for classifying images with a single stimulation.