Motion detecting method and apparatus

• US 7,813,530 B2
• Filed: 03/21/2007
• Issued: 10/12/2010
• Est. Priority Date: 03/22/2006
• Status: Active Grant

• Current Assignee: Nissan Motor Co., Ltd.
• Original Assignee: Nissan Motor Co., Ltd.
• Inventors: Watanabe, Seigo
• Primary Examiner(s): Bella; Matthew C
• Assistant Examiner(s): Nakhjavan; Shervin
• Application Number:
  US11/726,285
• Publication Number:
  US 20070230785A1
• Time in Patent Office: 1,301 Days
• Field of Search: 382/107, 382/190, 382/192, 382/194, 382/221, 348/148, 396/153
• US Class Current: 382/107
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments

Abstract

A motion detection apparatuses and method where a characteristic region is extracted from an image produced by an imaging device are described. A modified region is formed by modifying the characteristic region. Voting occurs so as to count up to increase a count value at a memory address corresponding to a location inside the modified region and to reset a count value at a memory address corresponding to a location outside the modified region. At least one of the moving direction and the moving speed of the modified region is calculated based on a difference among accumulated count values for adjacent locations. The position of the modified region can be calculated based on count values and locations in the modified region.

Claims

1. A motion detection apparatus, comprising:

an imaging device; and

a control unit operable to;

extract a characteristic region from an image produced by the imaging device;

form a modified region by modifying the characteristic region;

count up to increase a first count value at a first memory address corresponding to a location inside the modified region;

reset a second count value at a second memory address corresponding to a location outside the modified regions; and

calculate a position of the modified region based on count values and locations in the modified region by dividing a summation of products, each of the products obtained by multiplying each location in the modified region by a corresponding count value, by a summation of the corresponding count values of the locations in the modified region.

2. The apparatus according to claim 1 wherein the control unit is further operable to extract the characteristic region from the image produced by the imaging device by extracting a region in which an image density is higher than a predetermined threshold.

3. The apparatus according to claim 2 wherein the predetermined threshold is lower than a noise level of noise superposed on the image produced by the imaging device.

4. The apparatus according to claim 1 wherein the control unit is further operable to modify the characteristic region by calculating a gravity center position of the characteristic region and forming the modified region spreading in a range of pixels around the gravity center position as a center.

5. The apparatus according to claim 1 wherein the control unit is further operable to modify the characteristic region by modifying the characteristic region based on a speed of an object whose speed is to be detected.

6. The apparatus according to claim 1 wherein the controller is further operable to:

receive images produced by the imaging device in a form of consecutive frames; and

wherein a frame rate of the consecutive frames is based on an assumed moving speed of an object to be detected on the images.

7. The apparatus according to claim 1, further comprising:

calculate a moving direction and a moving speed of the modified region based on a difference among accumulated count values for adjacent locations.

8. A motion detection apparatus, comprising:

imaging means for forming an image of a scene around a mounting location;

characteristic region extracting means for extracting a characteristic region in the image;

characteristic region modifying means for forming a modified region by modifying the characteristic region, wherein the characteristic region modifying means includes;

means for calculating a gravity center position of the characteristic region by multiplying edge intensity values of pixels with the characteristic region by their respective coordinate locations to obtain a respective product, summing each respective product together to obtain a summation of products, summing the edge intensity values of the pixels within the characteristic region to obtain a summation of intensity values and dividing the summation of products by the summation of intensity values to obtain the gravity center position; and

means for forming the modified region using a range of pixels around the gravity center position;

voting means for increasing a first count value in a first memory address corresponding to a location inside the modified region and for resetting a second count value at a second memory address corresponding to a location outside the modified region; and

position calculating means for calculating a position of the modified region based on count values and locations in the modified region.

9. The apparatus according to claim 8 wherein the mounting location is a vehicle.

10. The apparatus according to claim 8 wherein the position calculating means comprises:

means for multiplying each location in the modified region by a corresponding count value to obtain a respective product;

means for summing each of the respective products together to obtain a summation of products;

means for summing the corresponding count values of the locations in the modified region to obtain a summation of count values; and

means for dividing the summation of products by the summation of count values to obtain the position.

11. The apparatus according to claim 10, further comprising:

moving speed calculating means for calculating a moving direction and a moving speed of the modified region based on a difference among count values accumulated at adjacent memory addresses by the voting means.

12. A motion detection method, comprising:

extracting a characteristic region from an input image;

modifying the characteristic region to form a modified region;

counting to increase or decrease a first count value at a first memory address corresponding to a location inside the modified region;

resetting a second count value at a second memory address corresponding to a location outside the modified region;

calculating at least one of a moving direction and a moving speed of the modified region based on a difference of count values accumulated at memory addresses for at least two adjacent locations; and

calculating a position of the modified region based on locations in the modified region and respective count values at the memory addresses by;

multiplying each location in the modified region by a corresponding count value to obtain a respective product;

summing each of the respective products together to obtain a summation of products;

summing the corresponding count values of the locations in the modified region to obtain a summation of count values; and

dividing the summation of products by the summation of count values to obtain the position.

13. The method according to claim 12 wherein extracting the characteristic region from the input image further comprises:

extracting a region in which an image density is higher than a predetermined threshold, the region defining the characteristic region.

14. The method according to claim 13 wherein modifying the characteristic region to form a modified region further comprises:

calculating a gravity center position of the characteristic region; and

forming the modified region using a range of pixels around the gravity center position.

15. The method according to claim 14, further comprising:

providing feedback from the moving speed of the modified region; and

modifying the characteristic region based on the moving speed.

16. The method according to claim 14, wherein calculating the gravity center position of the characteristic region comprises:

multiplying edge intensity values of pixels with the characteristic region by their respective coordinate locations to obtain a respective product;

summing each respective product together to obtain a summation of products;

summing the edge intensity values of the pixels within the characteristic region to obtain a summation of intensity values; and

dividing the summation of products by the summation of intensity values to obtain the gravity center position.

17. The method according to claim 12, further comprising:

producing the input image using an imaging device; and

setting the predetermined threshold lower than a noise level of noise superposed on the input image by the imaging device.

18. The method according to claim 12 wherein modifying the characteristic region to form a modified region further comprises:

calculating a gravity center position of the characteristic region; and

forming the modified region using a range of a plurality of pixels around the gravity center position.

19. The method according to claim 12, further comprising:

modifying the characteristic region based on a speed of an object whose speed is to be detected.

20. The method according to claim 19, further comprising:

receiving the input image as consecutive frames; and

setting a frame rate of the consecutive frames based on an assumed moving speed of the object whose speed is to be detected.

21. The method according to claim 12, further comprising:

receiving the input images as a series of consecutive frames; and

wherein extracting the characteristic region from the input image includes extracting a characteristic region from each of the series of consecutive frames.