OBJECT DETECTION USING DIFFERENCE OF IMAGE FRAMES

• US 20140233787A1
• Filed: 02/15/2013
• Published: 08/21/2014
• Est. Priority Date: 02/15/2013
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Mobile Communications AB (Sony Group Corp.)
• Inventors: Hellman, Erik, Westenius, Erik
• Granted Patent: US 9,014,428 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06V 20/46 Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
• CPC Class Codes: G06V 40/28 Recognition of hand or arm movements, e.g. recognition of deaf sign language static hand signs G06V40/113

Abstract

Object detection using a difference between image frames may include receiving a first image of a field of view, receiving a second image of the field of view, determining a difference between portions of the first image and corresponding portions of the second image, and declaring based on the difference between the portions of the first image and the corresponding portions of the second image that a specific object has been detected in the field of view.

Claims

1. An image capturing device for object detection, the image capturing device comprising:

a camera configured to obtain data representing a first image and a second image of objects in a field of view of the camera; and

an object detector including;

a difference calculator configured to receive the data representing the first image and the second image and determine difference between portions of the first image and corresponding portions of the second image, and

a classifier configured to declare, based on the difference between the portions of the first image and the corresponding portions of the second image, that a specific object has been detected in the field of view of the camera.

2. The image capturing device of claim 1, wherein

the difference calculator is configured to generate a difference image by performing a portion to portion comparison of the portions of the first image and the corresponding portions of the second image,

the difference image includes portions corresponding to the difference between the portions of the first image and the corresponding portions of the second image as determined by the comparison of the portions of the first image and the corresponding portions of the second image, and

the portions of the difference image are binarily set or unset based on a comparison of the difference between the portions of the first image and the corresponding portions of the second image and a threshold.

3. The image capturing device of claim 1, wherein the difference calculator is configured to perform a pixel by pixel comparison of pixels of the first image and corresponding pixels of the second image to determine difference between the pixels of the first image and the corresponding pixels of the second image.

4. The image capturing device of claim 3, wherein the difference calculator is configured to generate a difference image whose pixels correspond to the difference between the pixels of the first image and the corresponding pixels of the second image.

5. The image capturing device of claim 4, wherein the difference calculator is configured to binarily declare pixels of the difference image set or unset based on a comparison of the difference between the pixels of the first image and the corresponding pixels of the second image and a threshold.

6. The image capturing device of claim 4, comprising:

a data store including data corresponding to sample features of objects including features of the specific object,

wherein the classifier is configured to compare features in the difference image to the sample features to declare that the specific object has been detected in the field of view of the camera.

7. The image capturing device of claim 4, wherein the classifier is configured to perform an optical character recognition (OCR) of the difference image to recognize a pattern in the difference image corresponding to the specific object.

8. The image capturing device of claim 1, wherein the classifier is configured to declare based on a relatively large difference between the portions of the first image and the corresponding portions of the second image that movement of the object has been detected.

9. A method for object detection, the method comprising:

receiving a first image of a field of view;

receiving a second image of the field of view;

determining difference between portions of the first image and corresponding portions of the second image; and

declaring based on the difference between the portions of the first image and the corresponding portions of the second image that a specific object has been detected in the field of view.

10. The method of claim 9, wherein the determining the difference between the portions of the first image and the corresponding portions of the second image includes:

performing a pixel by pixel comparison of pixels of the first image and corresponding pixels of the second image to determine difference between the pixels of the first image and the corresponding pixels of the second image.

11. The method of claim 10, wherein the determining the difference between the portions of the first image and the corresponding portions of the second image includes:

generating a difference image whose pixels correspond to the difference between the pixels of the first image and the corresponding pixels of the second image.

12. The method of claim 11, wherein the determining the difference between the portions of the first image and the corresponding portions of the second image includes:

binarily declaring pixels of the difference image set or unset based on a comparison of the difference between the pixels of the first image and the corresponding pixels of the second image and a threshold.

13. The method of claim 11, wherein the declaring based on the difference between the portions of the first image and the corresponding portions of the second image that the specific object has been detected includes:

comparing features in the difference image to predetermined sample features of objects.

14. The method of claim 11, wherein the declaring based on the difference between the portions of the first image and the corresponding portions of the second image that the specific object has been detected includes:

performing an optical character recognition (OCR) of the difference image to recognize a pattern in the difference image corresponding to the specific object.

15. The method of claim 9, wherein the declaring based on the difference between the portions of the first image and the corresponding portions of the second image that the object has been detected in the field of view includes:

declaring based on a relatively large difference between the portions of the first image and the corresponding portions of the second image that movement has been detected.

16. The method of claim 15, wherein the declaring based on the relatively large difference between the portions of the first image and the corresponding portions of the second image that movement of has been detected includes:

comparing features in the difference image to predetermined sample features of objects to identify features in the difference image corresponding to specific movements or gestures.

17. The method of claim 9, wherein the determining the difference between the portions of the first image and the corresponding portions of the second image includes:

generating a difference image by performing a portion to portion comparison of the portions of the first image and the corresponding portions of the second image, wherein the difference image includes portions corresponding to the difference between the portions of the first image and the corresponding portions of the second image as determined by the comparison of the portions of the first image and the corresponding portions of the second image, and wherein the portions of the difference image are binarily set or unset based on a comparison of the difference between the portions of the first image and the corresponding portions of the second image and a threshold.

18. An object detector comprising:

a processor;

a data store operatively connected to the processor and including data representing a first image and a second image of a field of view;

a difference calculator configured to receive the data representing the first image and the second image and determine difference between portions of the first image and corresponding portions of the second image; and

a classifier configured to declare, based on the difference between the portions of the first image and the corresponding portions of the second image, that a specific object has been detected in the field of view.

19. The object detector of claim 18, wherein

the difference calculator is configured to generate a difference image by performing a portion to portion comparison of the portions of the first image and the corresponding portions of the second image,

the difference image includes portions corresponding to the difference between the portions of the first image and the corresponding portions of the second image as determined by the comparison of the portions of the first image and the corresponding portions of the second image, and

the portions of the difference image are binarily set or unset based on a comparison of the difference between the portions of the first image and the corresponding portions of the second image and a threshold.

20. The object detector of claim 18, wherein

the difference calculator is configured to:

perform a pixel by pixel comparison of pixels of the first image and corresponding pixels of the second image to determine difference between the pixels of the first image and the corresponding pixels of the second image, and

binarily declaring pixels of a difference image set or unset based on a comparison of the difference between the pixels of the first image and the corresponding pixels of the second image and a threshold, and

the classifier is configured to;

compare features in the difference image to predetermined sample features to declare that the specific object has been detected in the field of view.

21. The object detector of claim 18, wherein the classifier is configured to declare based on a relatively large difference between the portions of the first image and the corresponding portions of the second image at least one of:

movement has been detected in the field of view, or

a specific movement or gesture has been detected in the field of view.