User recognition apparatus and method

1. A user recognition apparatus comprising:

a visible-light image processing unit configured to acquire a visible-light image using a visible-light camera;

an infrared-light image processing unit configured to acquire an infrared-light image using an infrared-light camera;

a threshold value processing unit configured to remove a background of the infrared-light image using a brightness threshold, and generate a clipped infrared-light image;

an optical flow processing unit configured to calculate an optical flow based on the clipped infrared-light image and the visible-light image, and generate a clipped visible-light image based on the calculated optical flow, wherein the optical flow processing unit extracts feature points of the user based on the clipped infrared-light image, and maps the extracted user feature points to the visible-light image to calculate an optical flow; and

a visible light processing unit configured to recognize a user based on the clipped visible-light image.

2. The user recognition apparatus of claim 1 , wherein the visible light processing unit applies the clipped visible-light image to a visible-light algorithm to recognize the user, and determines a motion state of the user based on the recognition result.

3. The user recognition apparatus of claim 2 , wherein the optical flow processing unit generates depth information based on a vector length of the optical flow.

4. The user recognition apparatus of claim 1 , further comprising a synchronization unit configured to synchronize the infrared-light camera and the visible-light camera such that the infrared-light camera and the visible-light camera acquire images corresponding to a target at the same time.

5. The user recognition apparatus of claim 1 , further comprising a noise cancellation unit configured to cancel noises of the clipped infrared-light image and the visible-light image.

6. A user recognition method comprising:

acquiring an infrared-light image using an infrared-light camera;

acquiring a visible-light image using a visible-light camera;

removing a background of the infrared-light image using a brightness threshold, and generating a clipped infrared-light image;

calculating an optical flow based on the clipped infrared-light image and the visible-light image, wherein calculating the optical flow based on the clipped infrared-light image and the visible-light image comprises extracting feature points of the user based on the clipped infrared-light image and mapping the extracted user feature points to the visible-light image to calculate the optical flow;

generating a clipped visible-light image based on the calculated optical flow;

recognizing a user based on the clipped visible-light image.

7. The user recognition method of claim 6 , wherein the recognizing the user based on the clipped visible-light image comprises:

applying the clipped visible-light image to a visible-light algorithm to recognize the user; and

determining the motion state of the user based on the recognition result.

8. The user recognition method of claim 6 , wherein the calculating the optical flow based on the clipped infrared-light image and the visible-light image comprises generating depth information based on a vector length of the optical flow.

9. The user recognition method of claim 6 , wherein the infrared-light camera and the visible-light camera are synchronized to acquire images corresponding to a target at the same time.

10. The user recognition method of claim 6 , further comprising cancelling noises of the clipped infrared-light image and the visible-light image.