Image classification by brain computer interface

A method of classifying an image, comprising:

• presenting the image to an observer as a visual stimulus, while collecting neurophysiological signals from a brain of said observer;
• digitizing said neurophysiological signals to provide neurophysiological data;
• simultaneously processing the image and said neurophysiological data using a convolutional neural network (CNN) to identify a correlation between a computer vision detection of a target in the image and a neurophysiological event indicative of a detection of the target by said observer;
• and
• determining an existence of said target in the image based said identified correlation;
• wherein said CNN comprises a first convolutional neural subnetwork (CNS) configured for receiving and processing said neurophysiological data, a second CNS configured for receiving and processing the image, and a shared subnetwork having a neural network layer receiving and combining outputs from both said first CNS and said second CNS.