eduzhai > Applied Sciences > Engineering >

Decoding Visual Recognition of Objects from EEG Signals based on Attention-Driven Convolutional Neural Network

  • king
  • (0) Download
  • 20210507
  • Save

... pages left unread,continue reading

Document pages: 6 pages

Abstract: The ability to perceive and recognize objects is fundamental for theinteraction with the external environment. Studies that investigate them andtheir relationship with brain activity changes have been increasing due to thepossible application in an intuitive brain-machine interface (BMI). Inaddition, the distinctive patterns when presenting different visual stimulithat make data differentiable enough to be classified have been studied.However, reported classification accuracy still low or employed techniques forobtaining brain signals are impractical to use in real environments. In thisstudy, we aim to decode electroencephalography (EEG) signals depending on theprovided visual stimulus. Subjects were presented with 72 photographs belongingto 6 different semantic categories. We classified 6 categories and 72 exemplarsaccording to visual stimuli using EEG signals. In order to achieve a highclassification accuracy, we proposed an attention driven convolutional neuralnetwork and compared our results with conventional methods used for classifyingEEG signals. We reported an accuracy of 50.37 and 26.75 for 6-class and72-class, respectively. These results statistically outperformed otherconventional methods. This was possible because of the application of theattention network using human visual pathways. Our findings showed that EEGsignals are possible to differentiate when subjects are presented with visualstimulus of different semantic categories and at an exemplar-level with a highclassification accuracy; this demonstrates its viability to be applied it in areal-world BMI.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×