Monitoring camera device and related region-based motion detection method

• US 9,904,853 B2
• Filed: 12/20/2015
• Issued: 02/27/2018
• Est. Priority Date: 09/23/2015
• Status: Active Grant

• Current Assignee: Vatics Inc. (Kneron Inc.)
• Original Assignee: Vatics Inc. (Kneron Inc.)
• Inventors: Hsu, Te-Wei, Yeh, Chao-Chi
• Primary Examiner(s): Huang, Frank
• Application Number:
  US14/975,850
• Publication Number:
  US 20170084045A1
• Time in Patent Office: 800 Days
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 2207/20036 Morphological image processing
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
• 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 13/19604 involving reference image or background adaptation with time to compensate for changing conditions, e.g. reference image update on detection of light level change
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

A region-based motion detection method and a related monitoring camera device are applied to detect an object'"'"'s motion on a frame. The region-based motion detection method includes dividing a current frame into a plurality of first template arrays, dividing a first reference frame into a plurality of second template arrays, acquiring characteristic values of one of the first template arrays and a corresponding second template array, comparing the characteristic value of the foresaid first template array with the characteristic value of the corresponding second template array by matching process, and executing motion detection of the current frame according to a matching result.

Claims

1. A region-based motion detection method capable of detecting motion of an object displayed on a frame, the region-based motion detection method comprising:

detecting noise level of a current frame;

deciding array parameters of template arrays according to the noise level;

dividing the current frame into a plurality of first template arrays according to the array parameters;

dividing a first reference frame into a plurality of second template arrays according to the array parameters;

acquiring characteristic values of one of the plurality of first template arrays and a corresponding second template array of the plurality of second template arrays;

comparing the characteristic value of the foresaid first template array with the characteristic value of the corresponding second template array by matching process; and

executing motion detection of the object displayed on the current frame according to a result of the matching process.

2. The region-based motion detection method of claim 1, wherein the characteristic value is selected from a group consisting of an edge value, a chroma value, a datum variation value, a brightness value, a saturation value, and combination thereof.

3. The region-based motion detection method of claim 2, wherein the matching process optionally acquires a maximum value, a mean value and/or a weight value of the characteristic values for the motion detection.

4. The region-based motion detection method of claim 1, wherein the matching process utilizes value variation and/or phase variation of the characteristic values for the motion detection.

5. The region-based motion detection method of claim 1, wherein the array parameters are increased while the noise level is magnified.

6. The region-based motion detection method of claim 1, further comprising:

transforming the characteristic values by grouping process while the matching process is finished, wherein the region-based motion detection method utilizes the characteristic values transformed by the grouping process to execute the motion detection of the current frame.

7. The region-based motion detection method of claim 6, wherein a step of transforming the characteristic values by the grouping process while the matching process is finished comprises:

utilizing erosion calculation and dilation calculation of morphology to eliminate noise from the characteristic value.

8. The region-based motion detection method of claim 6, wherein low-pass filter, median filter and/or maximum filter is utilized to execute the grouping process.

9. The region-based motion detection method of claim 1, further comprising:

acquiring a plurality of weighted values according to a result of the motion detection;

applying the plurality of weighted values to the current frame and the first reference frame to generate a second reference frame; and

replacing the first reference frame by the second reference frame for next motion detection.

10. The region-based motion detection method of claim 9, wherein the plurality of weighted values comprise a first weighted value and a second weighted value varied according to the result of the motion detection, a product of the first weighted value and the current frame and a product of the second weighted value and the first reference frame are added to get the second reference frame.

11. The region-based motion detection method of claim 10, wherein the second weighted value is increased while an object motion on the current frame is decreased, and a sum of the first weighted value and the second weighted value equals 100%.

12. The region-based motion detection method of claim 9, wherein a step of acquiring the plurality of weighted values according to the result of the motion detection comprises:

acquiring the plurality of weighted values by lookup table technique and/or curve mapping technique.

13. A monitoring camera device, comprising:

a frame capturing unit adapted to capture a reference frame and a current frame respectively in different time; and

an operation processing unit electrically connected to the frame capturing unit and adapted to execute a region-based motion detection method of detecting noise level of the current frame, deciding array parameters of template arrays according to the noise level, dividing the current frame into a plurality of first template arrays according to the array parameters, dividing the reference frame into a plurality of second template arrays according to the array parameters, acquiring characteristic values of one of the plurality of first template arrays and a corresponding second template array of the plurality of second template arrays, comparing the characteristic value of the foresaid first template array with the characteristic value of the corresponding second template array by matching process, and executing motion detection of an object displayed on the current frame according to a result of the matching process.