Apparatus and method for checking gaze object

• US 9,400,552 B2
• Filed: 08/06/2013
• Issued: 07/26/2016
• Est. Priority Date: 12/12/2012
• Status: Active Grant

• Current Assignee: Hyundai Motor Company (Hyundai Motor Group)
• Original Assignee: Hyundai Motor Company (Hyundai Motor Group)
• Inventors: Park, Sung Min, Lee, Seok Beom, Ro, Hee Jin, Seok, Dong Hee
• Primary Examiner(s): Lao, Lun-Yi
• Assistant Examiner(s): Suteerawongsa, Jarurat
• Application Number:
  US13/960,390
• Publication Number:
  US 20140160011A1
• Time in Patent Office: 1,085 Days
• Field of Search: 345/156, 345 7- 8, 349/11
• US Class Current: 1/1
• CPC Class Codes: G06F 3/013 Eye tracking input arrangements G06F3/015 takes precedence

Abstract

An apparatus for checking a gaze object includes an eye detector configured to detect eyes from a captured image of a user who stares at objects while the objects displayed on a display move at certain frequencies, a frequency converter configured to check a motion of the detected eyes and convert the motion of the eyes into an eye moving frequency, and a discriminating unit configured to compare the eye moving frequency and moving frequencies of the objects and discriminate an object at which the user stares depending upon whether or not the eye moving frequency is identical with the moving frequencies of the objects.

Claims

1. An apparatus for checking a gaze object, the apparatus comprising:

a display controller configured to set unique moving frequencies for a plurality of warning lights displayed on a display of a vehicle and control the plurality of warning lights displayed on the display to move at the set unique moving frequencies;

an eye detector configured to detect eyes from a captured image of a user who stares at the display while the plurality of warning lights move at the set unique moving frequencies on the display;

a frequency converter configured to check a motion of the detected eyes and calculate an eye moving frequency corresponding to the motion of the eyes; and

a discriminator configured to receive the set unique moving frequencies to compare the eye moving frequency with the set unique moving frequencies and to discriminate a corresponding warning light at which the user stares depending upon whether or not the eye moving frequency is identical with the set unique moving frequencies.

2. The apparatus of claim 1, wherein the frequency converter performs a Fourier transform on the motion of the eye to calculate the eye moving frequency.

3. The apparatus of claim 1, wherein the frequency converter calculates the eye moving frequency corresponding to the motion of the eyes using any one algorithm of fast Fourier transform (FFT) and unequally spaced FFT.

4. The apparatus of claim 1, wherein the eye detector detects the eyes from the captured image using any one algorithm selected from the group consisting of Adaboost, Template matching, active appearance model (AAM) and active shape model (ASM).

5. The apparatus of claim 1, wherein the display controller controls the plurality of warning lights that the user stares at, to be turned off on the display according to a pre-discrimination result.

6. A method for checking a gaze object, the method comprising steps of:

setting unique moving frequencies for a plurality warning lights displayed on a display of a vehicle;

controlling the plurality of warning lights displayed on the display to move at the set unique moving frequencies;

detecting eyes from a captured image of a user who stares at display while the plurality of warning lights move at the set unique moving frequencies on the display;

checking a motion of the detected eyes and calculating an eye moving frequency corresponding to the motion of the eyes; and

comparing the eye moving frequency with the set unique moving frequencies and discriminating a corresponding warning light at which the user stares depending upon whether or not the eye moving frequency is identical with the set unique moving frequencies.

7. The method of claim 6, wherein the converting includes performing a Fourier transform on the motion of the eyes to calculate the eye moving frequency.

8. The method of claim 6, wherein the step of converting includes calculating the eye moving frequency corresponding to the motion of the eyes using any one algorithm of FFT and unequally spaced FFT.

9. The method of claim 6, wherein the step of detecting the eyes includes detecting the eyes from the image using any one algorithm selected from the group consisting of Adaboost, Template matching, active appearance model (AAM) and active shape model (ASM).

10. The method of claim 6, further comprising controlling the warning light that the user stares at, to be turned off on the display according to a discrimination result.