1 / 17

BLINK DETECTION AND TRACKING OF EYES FOR EYE LOCALISATION

BLINK DETECTION AND TRACKING OF EYES FOR EYE LOCALISATION. BY. LOPAMUDRA MOHAPATRA. (200199200 ). INTRODUCTION. What is blink detection? What is eye tracking? What is localization of eye?. WHY WE GO FOR EYE LOCALIZATION. Face normalization

tegan
Download Presentation

BLINK DETECTION AND TRACKING OF EYES FOR EYE LOCALISATION

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. BLINK DETECTION AND TRACKING OF EYES FOR EYE LOCALISATION BY LOPAMUDRA MOHAPATRA (200199200)

  2. INTRODUCTION • What is blink detection? • What is eye tracking? • What is localization of eye?

  3. WHY WE GO FOR EYE LOCALIZATION • Face normalization • Eye gaze based human computer interface. • For reading detection. • Security systems using the human iris for identification.

  4. PROPOSED PROTOCOL WHOLE METHOD THRESHHOLDING FRAMEDIFFERENCING EYE LOCALIZATION EYETRACKING

  5. THRESH HOLDING

  6. FRAME DIFFERENCING

  7. FRAME DIFFERENCING

  8. ALGORITHMS Steps in the blink detection (1) Obtain location of possible motion using Frame differencing. (2) Suitably thresh hold the motion regions and obtain blobs using morphological operation and connected components.

  9. ALGORITHM CONTD…... • (3) Remove unsuitable blobs that is either too big or too small or have incorrect width to height ratios to be considered as eyes. • (4) Repeat (1) to (3) until a suitable pair of blobs are found and mark their positions. • (5) Compute optical flow field in the blob regions • (6)Mark dominant direction of motion of blobs.

  10. Algorithm contd….. If the dominant motion is downward in a pair of blobs their positions are noted.These would represent eye closure during a blink. If the motion is not downward then steps (1) to (6) are repeated. (7) Repeat steps (1) to (6). (8)Discard blobs that are not suited near the location of the blobs found with downward motion.

  11. Algorithm contd • (8) Compute optical flow to ascertain if the dominant motion is upward with two ball remaining or repeat from step (7). • (9) If the dominant motion is upward, then classify the frame beginning from the frame where downward motion was detected to the frame where upward motion was detected as blink frames. If after downward motion no upward motion is detected upto 3 frames it is considered as no blinks. Process of blink detection is started from newframe.

  12. (10) The bounding boxes of the blobs where blink is deemed to have occurred is taken as eye detection. OPTICAL FLOW METHOD: • It allows for the differentiation between vertical eyelid movements during blinks and movement of eyeball and horizotal head movements. • EYE TRACKING : • After the location of eyes tracking is done by using KLT tracker.

  13. EYETRACKING • In eye tracking mainly there are 20 feature points are taken,which gives the more accuracy. • These feature pts are taken from the eye area and they are tracked in different places,and reinitialization is done.

  14. Results of eye tracking (a) eye region initialized (b) tracked eye regions to a movement just before blink.

  15. COMPUTATION SPEEDUP To speed up localization we need to speed up in - Optical Flow - Eye tracking. EXPERIMENTAL RESULT : (1)Optical flow: 10 sec (2) Tracking of eyes : 10 sec (3) reading image from disk: 13 sec

  16. CONCLUSION In this paper we have proposed an accurate and fast method for locating and tracking the eyes of a computer user situated in front of the monitor. By computing optical flow and using both the magnitude and direction of the flow vectors, we can differentiate blinking from the other motions. In this way our study completed.

  17. THANK YOU!!!

More Related