Motion detection device

• US 8,135,177 B2
• Filed: 09/26/2006
• Issued: 03/13/2012
• Est. Priority Date: 09/27/2005
• Status: Expired due to Fees

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Benten, Harold Gerardus Pieter Hendriku
• Primary Examiner(s): Tabatabai, Abolfazl
• Application Number:
  US12/067,962
• Publication Number:
  US 20080205710A1
• Time in Patent Office: 1,995 Days
• Field of Search: 382/103, 382/104, 382/107, 382/236, 348/14.1, 348/97, 348/154, 348/155, 348/352, 348/407.1, 348/413.1, 348/430, 348/715, 348/169, 348/208.1, 348/208.2, 348/208.16, 348/451, 375/240.1, 375/240.16, 396/153
• US Class Current: 382/107
• CPC Class Codes: G06T 7/238 using non-full search, e.g. three-step search
• CPC Class Codes: G08G 1/017 identifying vehicles G08G1/015, G08G1/054 take precedence
• CPC Class Codes: G08G 1/052 with provision for determining speed or overspeed speed measuring in general G01P

Abstract

The invention refers to a device, a method and a computer program product for extracting motion information from a sequence of video frames. Existing solutions for extracting motion information from sequence of video frames need a massive computing power which make it difficult and expensive to implement a real-time system. It is therefore an object of the invention to simplify such a device and to provide a real-time embedded system. It is suggested to provide a device comprising a digital video camera 1. The video camera 1 includes a processing unit 3 for processing video frames grabbed by the video camera 1. The processing uses a 3D recursive search block matching algorithm to extract the motion information from the video frames. The device can be used for traffic surveillance applications, e.g. for determining the speed of vehicles on the streets and roads.

Claims

1. A device for extracting motion information from unencoded video frames, comprising:

a digital video camera for grabbing the unencoded video frames,

a processing unit for processing the unencoded video frames provided by the video camera,

the processing unit being adapted to use a recursive search block algorithm to determine whether the unencoded video frames show an object is moving or whether the unencoded video frames show a person is moving;

wherein the processing unit is further adapted to determine the velocity of the object or the person, based on a frame rate of the unencoded video frames and a conversion factor for converting a pixel velocity into a physical velocity.

2. The device according to claim 1, wherein the processing unit is adapted to determine simultaneously the velocity of a multitude of objects or persons.

3. The device according to claim 1, wherein the processing unit is located in the housing of the video camera.

4. The device according to claim 1, wherein the processing unit is implemented as a real-time system.

5. The device according to claim 1, wherein the processing unit is adapted to indicate the position of the moving object or the moving person.

6. The device according to claim 1, wherein the processing unit is adapted to carry out an object recognition.

7. The device according to claim 1, wherein the processing unit is adapted to carry out a number plate recognition.

8. A method for extracting motion information from unencoded video frames, comprising:

grabbing, by a digital video camera, the unencoded video frames,

processing, by a processing unit, the unencoded digital video frames, the processing being carried out by using a recursive search block algorithm to determine whether the unencoded video frames show an object is moving or whether the unencoded video frames show a person is moving; and

determining the velocity of the object or the person, based on a frame rate of the unencoded video and a conversion factor for converting a pixel velocity into a physical velocity.

9. The method according to claim 8, wherein the velocity of a multitude of objects or persons are determined simultaneously.

10. The method according to claim 8, wherein surveillance applications are used for detecting an intrusion into a building or into premises.

11. The method according to claim 8, wherein a road-rule enforcement camera is used.

12. A non-transitory computer readable medium, having thereon a computer program, when executed, for carrying out acts comprising:

grabbing the unencoded video frames; and

processing the unencoded digital video frames, the processing being carried out by using a recursive search block algorithm to determine whether the unencoded video frames show an object is moving or whether the unencoded video frames show a person is moving; and

determining the velocity of the object or the person, based on a frame rate of the unencoded video frames and a conversion factor for converting a pixel velocity into a physical velocity.