Hough transform based motion detection image recording system and method

• US 20020071596A1
• Filed: 12/03/2001
• Published: 06/13/2002
• Est. Priority Date: 12/11/2000
• Status: Active Grant

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Estevez, Leonardo W.
• Granted Patent: US 7,035,434 B2
• Time in Patent Office: Days
• US Class Current: 382/107
• CPC Class Codes: G06T 7/262 using transform domain methods, e.g. Fourier domain methods

Abstract

A Hough based motion detection image recording system includes an image sensor operable to view a surveillance site and output real-time image data corresponding to the site. The system also includes a processor operable to receive the image data from the image sensor and a memory coupled to the processor. The system includes a transformation application residing in the memory and executable by the processor. The transformation application is operable to detect a real-time feature segment associated with the site from the image data. The system further includes an analysis engine residing in the memory and executable by the processor. The analysis engine is operable to compare the real-time feature segment with a baseline feature segment and initiate recording of the image data if the feature segment comparison exceeds a predetermined criteria.

Claims

1. A motion detection image recording system, comprising:

an image sensor operable to view a surveillance site and output real-time image data corresponding to the site;

a processor operable to receive the image data from the image sensor;

a memory coupled to the processor;

a transformation application residing in the memory and executable by the processor, the transformation application operable to detect a real-time feature segment associated with the site from the image data; and

an analysis engine residing in the memory and executable by the processor, the analysis engine operable to compare the real-time feature segment with a baseline feature segment and initiate recording of the image data if the feature segment comparison exceeds a predetermined criteria.

2. The system of claim 1, wherein the transformation application is further operable to generate a histogram of the real-time feature segment, and wherein the analysis engine is operable to compare the histogram of the real-time feature segment with a histogram of the baseline feature segment.

3. The system of claim 1, wherein the analysis engine is operable to identify an occlusion of the real-time feature segment relative to the baseline feature segment and initiate recording of the scene in response to identifying the occlusion.

4. The system of claim 1, wherein the image sensor comprises an analog camera, and further comprising an analog-to-digital converter operable to receive analog image data from the analog camera and transmit digital image data to the processor.

5. The system of claim 1, wherein the transformation application comprises a Hough routine operable to extract the real-time feature segment from the image data.

6. The system of claim 1, wherein the processor is further operable to receive feature segment criteria, and wherein the analysis engine initiates recording of the image data if the feature segment comparison exceeds the feature segment criteria.

7. The system of claim 1, wherein the feature segment criteria comprises an occlusion ratio, and wherein the analysis engine initiates recording of the image data if an occlusion of the real-time feature segment exceeds the occlusion ratio relative to the baseline feature segment.

8. A method for motion detection image recording, comprising:

generating a baseline feature segment associated with a surveillance site;

receiving real-time image data corresponding to the surveillance site;

detecting a real-time feature segment associated with the surveillance site from the real-time image data;

comparing the real-time feature segment with the baseline feature segment; and

recording the real-time image data if the comparison of the real-time feature segment with the baseline feature segment exceeds a feature segment criteria.

9. The method of claim 8, wherein detecting a real-time feature segment comprises transforming the real-time image data using a Hough routine to extract the real-time feature segment from the real-time image data.

10. The method of claim 8, wherein generating a baseline feature segment comprises:

receiving baseline image data corresponding to the surveillance site;

detecting a plurality of feature segments corresponding to the surveillance site from the baseline image data; and

identifying one of the plurality of feature segments as the baseline feature segment.

11. The method of claim 8, wherein comparing the real-time feature segment comprises determining whether an occlusion is present in the real-time feature segment relative to the baseline feature segment, and wherein recording the real-time image data comprises recording the real-time image data if the occlusion exceeds the feature segment criteria.

12. The method of claim 8, wherein receiving real-time image data comprises:

generating analog image data via an analog camera;

converting the analog image data to digital image data; and

transmitting the digital image data to a processor to detect the real-time feature segment.

13. The method of claim 8, wherein generating a baseline feature segment comprises:

receiving baseline image data corresponding to the surveillance site;

detecting a plurality of feature segments associated with the surveillance site from the baseline image data; and

identifying a subset of the plurality of feature segments as the baseline feature segment.

14. The method of claim 8, wherein comparing the real-time feature segment with the baseline feature segment comprises:

generating a baseline histogram associated with the baseline feature segment;

generating a real-time histogram associated with the real-time feature segment; and

comparing the baseline histogram to the real-time histogram.

15. The method of claim 8, wherein comparing the real-time feature segment to the baseline feature segment comprises:

determining a length of the baseline feature segment;

determining a length of the corresponding real-time feature segment; and

determining whether a difference between the length of the baseline feature segment and the length of the real-time feature segment exceeds the feature segment criteria.

17. The method of claim 16, further comprising:

receiving feature segment criteria;

determining whether the occlusion exceeds the feature segment criteria; and

wherein recording comprises recording the real-time image data if the occlusion exceeds the feature segment criteria.

18. The method of claim 16, wherein detecting the real-time feature segments comprises applying a Hough routine to the real-time image data to extract the real-time feature segments from the real-time image data.

19. The method of claim 16, further comprising:

generating a baseline histogram corresponding to the baseline feature segments;

generating a real-time histogram corresponding to the real-time feature segments; and

comparing the baseline histogram with the real-time histogram to determine whether an occlusion is present in one or more of the real-time feature segments.

20. The method of claim 16, further comprising recording the real-time image data if a quantity of the real-time feature segments exceeds a quantity of the baseline feature segments.

16. A method for image recording, comprising:

generating baseline feature segments corresponding to a surveillance site;

receiving real-time image data corresponding to the surveillance site;

generating real-time feature segments associated with the surveillance site from the real-time image data;

determining whether an occlusion is present in one or more of the real-time feature segments; and

recording the real-time image data if the occlusion is present in one or more of the real-time feature segments.