Method for detecting moving objects by comparing video images

• US 6,954,225 B2
• Filed: 04/11/2002
• Issued: 10/11/2005
• Est. Priority Date: 04/13/2001
• Status: Active Grant

• Current Assignee: Huper Laboratories Co., Ltd.
• Original Assignee: Huper Laboratories Co., Ltd.
• Inventors: Chen, Hsiao-Ping
• Primary Examiner(s): Ho, Tuan
• Assistant Examiner(s): Jelinek, Brian
• Application Number:
  US10/063,322
• Publication Number:
  US 20020180870A1
• Time in Patent Office: 1,279 Days
• Field of Search: 348/155, 348/152, 348/143, 348/360, 382/107, 382/236, 382/274, 382/192, 382/173, 382/103, 340/567, 340/511
• US Class Current: 348/207.99
• CPC Class Codes: G06T 7/254 involving subtraction of images
• CPC Class Codes: H04N 5/144 Movement detection for video coding H04N19/503; analysis of motion in general G06T7/20

Abstract

A method detects activity within a fixed view by comparing first and second images. The first and second images respectively include a plurality of blocks, and each of the blocks having a plurality of pixels. The method including: calculating an average value A for each block of the second image, each of the average values A is equal to an average of differences between pixels of one of the blocks of the second image and pixels of a corresponding block of the first image; calculating a mean difference B between average pixel values of the first and second images; calculating a difference V between average value A and mean difference B for each block of the second image, and selecting blocks of the second image that have corresponding differences V greater than a threshold value; and determining whether there is any object moving within the fixed view according to the selected blocks.

Claims

1. A method for detecting activity within a fixed view by using a monitor system, the monitor system comprising a camera for filming the fixed view at a predetermined camera speed and a calculator for detecting the activity within the fixed view according to a first image and a second image generated by the camera, the second image being generated within a predetermined time interval after the first image is generated, each of the images generated by the camera comprising a plurality of blocks, each of the blocks comprising a plurality of pixels, the method comprising:

calculating an average value A for each block of the second image, each of the average values A is equal to an average of differences between pixels of one of the blocks of the second image and pixels of a corresponding block of the first image;

calculating a mean difference B between an average value of pixels of the first image and an average value of pixels of the second image;

calculating a difference V between the average value A and the mean difference B for each block of the second image, and selecting blocks of the second image that have corresponding differences V greater than a threshold value; and

determining whether there is any object moving within the fixed view according to the selected blocks.

2. The method of claim 1 further comprising:

selecting a predetermined percentage of the blocks of the second image, wherein all of the average values A of the selected blocks are not greater than any average value A of the un-selected blocks;

calculating a mean value C of the average values A of the selected blocks; and

selecting blocks of the second image that have corresponding absolute values |A−

C| greater than the threshold value when the differences V are calculating.

3. The method of claim 2 further comprising:

calculating a difference mean value D that is equal to an average of differences between the average values A and the mean value C; and

selecting blocks of the second image that have corresponding absolute values |(A C)/D| greater than the threshold value when the differences V are calculating.

4. The method of claim 1 further comprising:

calculating a difference mean value F that is equal to an average of differences between the average values A and the mean difference B; and

selecting blocks of the second image that have corresponding absolute values |(A B)/F| greater than the threshold value when the differences V are calculating.

5. The method of claim 1 wherein each of the pixels comprises a plurality of image parameters, and the average values A and the mean difference B are calculated according to at least one of the image parameters.

6. The method of claim 5 wherein the image parameters include brightness, gray level, and color temperature.

7. The method of claim 1 wherein each of the pixels comprises a plurality of image parameters, and a representation value of each pixel can be obtained according to the image parameters, the average values A and the mean difference B being calculated according to the representation value of each pixel.

8. The method of claim 7 wherein the image parameters include brightness, gray level, and color temperature.

9. The method of claim 1 further comprising:

determining a moving area according to the selected blocks of the second image and a distribution rule for eliminating abnormal distributed blocks from the selected blocks.

10. The method of claim 1 wherein when any moving object is detected within the fixed view by the monitor system, the monitor system continues comparing the images generated by the camera to trace the moving object.

11. A method for detecting activity within a fixed view by using a monitor system, the monitor system comprising a camera for filming the fixed view at a predetermined camera speed and a calculator for detecting the activity within the fixed view according to a first image and a second image generated by the camera, the second image being generated within a predetermined time interval after the first image is generated, each of the images generated by the camera comprising a plurality of blocks, each of the blocks comprising a plurality of pixels, the method comprising:

selecting a detection area within the second image;

calculating an average value A for each block within the detection area, each of the average values A is equal to an average of differences between pixels of one of the blocks of the second image within the detection area and pixels of a corresponding block of the first image;

calculating a mean difference B between the average value of pixels of the second image within the detection area and the average value of corresponding pixels of the first image;

calculating a difference V between the average value A and the mean difference B for each block of the second image within the detection area, and selecting blocks of the second image within the detection area that have corresponding differences V greater than a threshold value; and

determining whether there is any object moving within a field of the fixed view that corresponds to the detection area according to the selected blocks.

12. The method of claim 11 further comprising:

selecting a predetermined percentage of the blocks within the detection area, wherein all of the average values A of the selected blocks are not greater than any average value A of the un-selected blocks;

calculating a mean value C of the average values A of the selected blocks; and

selecting blocks that have corresponding absolute values |A−

C| greater than the threshold value within the detection area when the differences V are calculating.

13. The method of claim 12 further comprising:

calculating a difference mean value D that is equal to an average of differences between the average values A and the mean value C; and

selecting blocks that have corresponding absolute values |(A C)/D| greater than the threshold value within the detection area when the differences V are calculating.

14. The method of claim 11 further comprising:

calculating a difference mean value F that is equal to an average of differences between the average values A and the mean difference B; and

selecting blocks that have corresponding absolute values |(A B)/F| greater than the threshold value within the detection area when the differences V are calculating.

15. The method of claim 11 wherein each of the pixels comprises a plurality of image parameters, and the average values A and the mean difference B are calculated according to at least one of the image parameters.

16. The method of claim 15 wherein the image parameters include brightness, gray level, and color temperature.

17. The method of claim 11 wherein each of the pixels comprises a plurality of image parameters, and a representation value of each pixel can be obtained according to the image parameters, the average values A and the mean difference B being calculated according to the representation values of the pixels within the detection area.

18. The method of claim 17 wherein the image parameters include brightness, gray level, and color temperature.

19. The method of claim 11 further comprising:

determining a moving area according to the selected blocks within the detection area and a distribution rule for eliminating abnormal distributed blocks from the selected blocks.

20. The method of claim 11 wherein when any moving object is detected in the field that corresponds to the detection area by the monitor system, the monitor system continues comparing the images generated by the camera to trace the moving object.