Object recognition apparatus and object recognition method

• US 8,861,784 B2
• Filed: 05/05/2010
• Issued: 10/14/2014
• Est. Priority Date: 05/11/2009
• Status: Active Grant

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Tojo, Hiroshi
• Primary Examiner(s): GORADIA, SHEFALI DINESH
• Application Number:
  US12/774,207
• Publication Number:
  US 20100284568A1
• Time in Patent Office: 1,623 Days
• Field of Search: 382/100, 382/103, 382/107, 382/190
• US Class Current: 382/103
• CPC Class Codes: G06V 40/166 using acquisition arrangements
• CPC Class Codes: G06V 40/167 using comparisons between temporally consecutive images

Abstract

An object recognition apparatus recognizes an object from video data for a predetermined time period generated by a camera, analyzes the recognition result, and determines a minimum size and moving speed of faces of the video image recognized from the received frame image. Then, the object recognition apparatus determines a lower limit value of a frame rate and resolution from the determined minimum size and moving speed of the faces.

Claims

1. An object recognition apparatus comprising:

an imaging unit configured to capture a moving image based on a frame rate to generate video data;

an object recognition unit configured to recognize an object from the video data generated by the imaging unit;

a recognition result analysis unit configured to estimate a positional displacement between frames for estimating a moving speed of the object recognized by the object recognition unit; and

a determination unit configured to determine a lower limit value of the frame rate to be used for generating succeeding video data by the imaging unit based on the moving speed estimated from the positional displacement between frames estimated by the recognition result analysis unit.

2. The object recognition apparatus according to claim 1,

wherein the object recognition unit recognizes the object from the video data for a predetermined time period generated by the imaging unit, and

wherein the determination unit determines the frame rate of the imaging unit based on the moving speed of the object in the video data for the predetermined time estimated by the recognition result analysis unit.

3. The object recognition apparatus according to claim 1,

wherein the recognition result analysis unit further estimates a size of the object recognized in the video data from the object recognition unit, and

wherein the determination unit further determines the resolution of the video data generated by the imaging unit based on the size of the object estimated by the recognition result analysis unit.

4. The object recognition apparatus according to claim 3, wherein the determination unit determines a lower limit value of the resolutions based on a minimum size of the object estimated by the recognition result analysis unit.

5. The object recognition apparatus according to claim 1, further comprising a designating unit configured to designate a recognition target region in the video data,

wherein the determination unit determines the frame rate based on an object recognition result in the recognition target region designated by the designating unit.

6. The object recognition apparatus according to claim 1, further comprising a setting unit configured to set a highest value available as an initial value of the frame rate.

7. An object recognition method comprising:

capturing a moving image based on a frame rate to generate video data;

recognizing an object from the generated video data;

estimating a positional displacement between frames for estimating a moving speed of the recognized object; and

determining a lower limit value of the frame rate to be used for generating succeeding video data, based on the estimated moving speed from the estimated positional displacement between frames of the recognized object.

8. A non-transitory computer-readable storage medium storing a program that causes a computer to execute an object recognition method, the method comprising:

capturing a moving image based on a frame rate to generate video data;

recognizing an object from the generated video data;

estimating a positional displacement between frames for estimating a moving speed of the recognized object; and

determining a lower limit value of the frame rate to be used for generating succeeding video data, based on the estimated moving speed from the estimated positional displacement between frames of the recognized object.