Visual attention distractor insertion for improved EEG RSVP target stimuli detection

• US 9,489,732 B1
• Filed: 03/12/2014
• Issued: 11/08/2016
• Est. Priority Date: 12/21/2010
• Status: Active Grant

• Current Assignee: HRL Laboratories LLC (The Boeing Co.)
• Original Assignee: HRL Laboratories LLC (The Boeing Co.)
• Inventors: Khosla, Deepak, Martin, Kevin R., Huber, David J.
• Primary Examiner(s): Entezari, Michelle
• Application Number:
  US14/206,421
• Time in Patent Office: 972 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: A61B 5/378 Visual stimuli
• CPC Class Codes: G06F 2218/00 Aspects of pattern recognition specially adapted for signal processing

Abstract

Described is a system for improved electroencephalograph (EEG) rapid serial visual presentation (RSVP) target stimuli detection through visual attention distractor insertion. A first RSVP sequence is created comprising a set of image chips. The image chips are a combination of target images containing target events and non-target images containing comment events. A number of visual attention distractors to optimize target event detection is determined, and the determined number of visual attention distractors is inserted into the first RSVP sequence to generate a second RSVP sequence. The second RSVP sequence is reordered to generate a third RSVP sequence. The third RSVP sequence is presented to a user, and an EEG signal is received from the user. Finally, the EEG signal is decoded to identify a true target event via a P300 detection in the EEG signal.

Claims

1. A system for improved target stimuli detection, the system comprising:

one or more processors and a non-transitory computer-readable medium having executable instructions encoded thereon such that when executed, the one or more processors perform operations of;

creating a first rapid serial visual presentation (RSVP) sequence, wherein the first RSVP sequence comprises a set of image chips, wherein the set of image chips comprises a combination of target images containing at least one target event and non-target images containing common events;

determining a number of visual attention distractors to optimize target event detection;

wherein the visual attention distractors are visually similar to the non-target images containing common events obtained during creation of the first RSVP sequence;

inserting the determined number of visual attention distractors into the first RSVP sequence to generate a second RSVP sequence;

reordering the second RSVP sequence to generate a third RSVP sequence;

presenting the third RSVP sequence to a user and receiving an electroencephalograph (EEG) signal from the user; and

decoding the EEG signal to identify a true target event via a P300 detection in the EEG signal.

2. The system as set forth in claim 1, wherein determining the number of visual attention distractors is based on establishing an optimal ratio of a probability of target events to common events for a maximum P300 detection in the EEG signal.

3. The system as set forth in claim 2, wherein the number of visual attention distractors, K, is determined according to the following:

K=[(R_E/R_I)−

1]*N if R_E>

R_I;

=0 otherwise,

where N denotes a number of image chips in the first RSVP sequence, E represents a number of expected target events in the first RSVP sequence;

P denotes a probability of detection of a target event in the first RSVP sequence by a cognitive algorithm, R_I represents an ideal ratio of target events to common events for a maximum P300 detection in the EEG signal, and an expected ratio of target events to common events, R_E, is equal to (P*E)/N.

4. The system as set forth in claim 3, wherein reordering the second RSVP sequence maximizes a likelihood of detecting a true target event and minimizes a likelihood of detecting false alarms.

5. A computer-implemented method for improved target stimuli detection, comprising an act of:

causing one or more processors to execute instructions stored on a non-transitory memory such that upon execution, the data processor performs operations of;

creating a first rapid serial visual presentation (RSVP) sequence, wherein the first RSVP sequence comprises a set of image chips, wherein the set of image chips comprises a combination of target images containing at least one target event and non-target images containing common events;

determining a number of visual attention distractors to optimize target event detection;

wherein the visual attention distractors are visually similar to the non-target images containing common events obtained during creation of the first RSVP sequence;

inserting the determined number of visual attention distractors into the first RSVP sequence to generate a second RSVP sequence;

reordering the second RSVP sequence to generate a third RSVP sequence;

presenting the third RSVP sequence to a user and receiving an electroencephalograph (EEG) signal from the user; and

decoding the EEG signal to identify a true target event via a P300 detection in the EEG signal.

6. The method as set forth in claim 5, wherein determining the number of visual attention distractors is based on establishing an optimal ratio of a probability of target events to common events for a maximum P300 detection in the EEG signal.

7. The method as set forth in claim 6, wherein the number of visual attention distractors, K, is determined according to the following:

K=[(R_E/R_I)−

1]*N if R_E>

R_I;

=0 otherwise,

where N denotes a number of image chips in the first RSVP sequence, E represents a number of expected target events in the first RSVP sequence;

P denotes a probability of detection of a target event in the first RSVP sequence by a cognitive algorithm, R_I represents an ideal ratio of target events to common events for a maximum P300 detection in the EEG signal, and an expected ratio of target events to common events, R_E, is equal to (P*E)/N.

8. The method as set forth in claim 7, wherein reordering the second RSVP sequence maximizes a likelihood of detecting a true target event and minimizes a likelihood of detecting false alarms.

9. A computer program product for improved target stimuli detection, the computer program product comprising computer-readable instructions stored on a non-transitory computer-readable medium that are executable by a computer having a processor for causing the processor to perform operations of:

creating a first rapid serial visual presentation (RSVP) sequence, wherein the first RSVP sequence comprises a set of image chips, wherein the set of image chips comprises a combination of target images containing at least one target event and non-target images containing common events;

determining a number of visual attention distractors to optimize target event detection;

wherein the visual attention distractors are visually similar to the non-target images containing common events obtained during creation of the first RSVP sequence;

inserting the determined number of visual attention distractors into the first RSVP sequence to generate a second RSVP sequence;

reordering the second RSVP sequence to generate a third RSVP sequence;

presenting the third RSVP sequence to a user and receiving an electroencephalograph (EEG) signal from the user; and

decoding the EEG signal to identify a true target event via a P300 detection in the EEG signal.

10. The computer program product as set forth in claim 9, wherein determining the number of visual attention distractors is based on establishing an optimal ratio of a probability of target events to common events for a maximum P300 detection in the EEG signal.

11. The computer program product as set forth in claim 10, wherein the number of visual attention distractors, K, is determined according to the following:

K=[(R_E/R_I)−

1]*N if R_E>

R_I;

=0 otherwise,

where N denotes a number of image chips in the first RSVP sequence, E represents a number of expected target events in the first RSVP sequence;

P denotes a probability of detection of a target event in the first RSVP sequence by a cognitive algorithm, R_I represents an ideal ratio of target events to common events for a maximum P300 detection in the EEG signal, and an expected ratio of target events to common events, R_E, is equal to (P*E)/N.

12. The computer program product as set forth in claim 11, wherein reordering the second RSVP sequence maximizes a likelihood of detecting a true target event and minimizes a likelihood of detecting false alarms.