Apparatus and a method for detecting motion within an image sequence

• US 6,628,805 B1
• Filed: 05/05/2000
• Issued: 09/30/2003
• Est. Priority Date: 06/17/1996
• Status: Expired due to Term

• Current Assignee: SRI International, Inc.
• Original Assignee: Sarnoff Corporation (SRI International, Inc.)
• Inventors: Burt, Peter Jeffrey, Hansen, Michael Wade
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): Lu, Tom Y.
• Application Number:
  US09/565,149
• Time in Patent Office: 1,243 Days
• Field of Search: 382/104, 382/107, 382/260, 382/278, 348/699, 348/95, 348/118, 701/207, 701/300, 356/375
• US Class Current: 382/107
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 7/248 involving reference images or patches
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G08B 13/19602 Image analysis to detect motion of the intruder, e.g. by frame subtraction
• CPC Class Codes: G08B 13/19643 wherein the cameras play different roles, e.g. different resolution, different camera type, master-slave camera
• CPC Class Codes: G08B 13/19652 Systems using zones in a single scene defined for different treatment, e.g. outer zone gives pre-alarm, inner zone gives alarm
• CPC Class Codes: H04N 5/145 Movement estimation for video coding H04N19/51

Abstract

Apparatus and a concomitant method of identifying the direction of motion within a scene that is represented by a sequence of images, e.g., a series of frames in a video sequence. The apparatus contains a flow field generator, a flow field segmentor, and alarm detector and an alarm processor. The flow field generator processes the sequence of images and generates a substantially distortionless flow field representing the motion within a scene. The flow field is a vector representation of the motion of the scene that represents both the magnitude and the direction of the motion. The flow field is generated by correlating at least two frames in the sequence of images. This flow field is analyzed by a flow field segmentor to determine the magnitude and direction of motion with the scene and segment the motion information from the static portions of the scene. An alarm detector then processes the motion information to determine if an alarm should be generated based upon the motion information. The alarm processor activates an appropriate alarm response, e.g., illuminate warning lamps, sound a warning alarm, lock doors, alert security personnel, and the like.

Claims

1. Apparatus for detecting motion within a scene that is represented by a sequence of images, comprising:

a flow field generator for processing said sequence of images to produce a motion flow field, wherein said flow field generator comprises an image preprocessor and a flow field processor, wherein said image preprocessor comprises;

an image selector for selecting four images from said sequence of images;

a first subtractor for subtracting a first image from a second image to create a first difference image;

a second subtractor for subtracting a third image from a fourth image to create a second difference image;

a first shaped filter for filtering said first difference image to produce a first filtered image; and

a second shaped filter for filtering said second difference image to produce a second filtered image;

a flow field segmentor, connected to said flow field generator, for extracting said motion flow field from said sequence of images; and

an alarm detector, connected to said flow field segmentor, for analyzing said motion flow field to determine if said motion flow field meets an alarm criteria.

2. The apparatus of claim 1 wherein said first and second filters are shaped filters.

3. The apparatus of claim 1 wherein said flow field processor further comprises:

a first image shifter for selectively shifting a first filtered image;

a second image shifter for selectively shifting a second filtered Image;

a multiplier for multiplying said first and second shifted filtered images with one another;

an integrator, connected to said multiplier, for integrating a series of multiplied images;

a flow field estimator, connected to said integrator, for generating said motion flow field from said integrated images.

4. The apparatus of claim 1 wherein said flow field generator further comprises a confidence value generator for producing a confidence value for each flow field vector in said motion flow field.

5. The apparatus of claim 1 further comprising means, connected to said flow field segmentor, for adjusting the sensitivity of the flow field generator in response to analyzing the segmented flow field generated by the flow field segmentor.

6. The apparatus of claim 5 wherein said sensitivity adjusting means controls the image selector.

7. The apparatus of claim 5 wherein said sensitivity adjusting means controls the flow field estimator.

8. A method for detecting motion within a scene that is represented by a sequence of images, comprising the steps of:

selecting four images from said sequence of images;

subtracting a first image from a second image to create a first difference Image;

subtracting a third image from a fourth image to create a second difference image;

filtering said first difference image to produce a first filtered image; and

filtering said second difference image to produce a second filtered image;

generating a motion flow field from said first and second filtered images; and

analyzing said motion flow field to determine if said motion flow field meets an alarm criteria.

9. The method of claim 8 wherein said first filtered image is produced by shaped filtering said first difference image and said second filtered image is produced by shaped filtering said second difference image.

10. The method of claim 8 wherein said processing step further comprises the steps of:

selectively shifting a first filtered image;

selectively shifting a second filtered image;

multiplying said first and second shifted filtered Images with one another;

integrating a series of multiplied images;

generating said motion flow field from said integrated images.

11. The method of claim 8 wherein said generating step further comprises the step of producing a confidence value for each flow field vector in said motion flow field.

12. The method of claim 11 further comprising the step of adjusting the sensitivity used to generate said flow field in response to analyzing the extracted flow field.