Flare Monitoring

• US 20090046172A1
• Filed: 08/14/2007
• Published: 02/19/2009
• Est. Priority Date: 08/14/2007
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Gelinboina, Mahesh K.., Rao, Venkatagiri S., Venkoparao, Vijendran G., Hota, Rudra N.
• Granted Patent: US 7,876,229 B2
• Time in Patent Office: Days
• US Class Current: 348/226.100
• CPC Class Codes: G06F 18/2411 based on the proximity to a decision surface, e.g. support vector machines
• CPC Class Codes: G06V 10/56 relating to colour
• CPC Class Codes: G06V 10/764 using classification, e.g. of video objects

Abstract

A computer assisted method identifies characteristics of a flare. A digital image of the flare is obtained, such as by a color digital video camera and a region of interest likely containing the flare is identified. Flare pixels within the region of interest are identified and color information is extracted from such flare pixels. The extracted color information is associated to characterizations of the flare, and an operator is provided an alert when such characterizations indicate an abnormal flare

Claims

1. A computer assisted method of identifying characteristics of a flare, the method comprising:

obtaining a digital image of the flare;

identifying a region of interest of said captured image;

identifying flare pixels within said region of interest;

extracting color information from the flare pixels;

associating the extracted color information to characterizations of said flare; and

providing an operator alert when such characterizations indicate an abnormal flare.

2. The method of claim 1 wherein identification of flare pixels is executed by change detection with background model.

3. The method of claim 1 wherein color information is extracted in a color space selected from the group consisting of RGB, HSV, YCbCr, and Normalized RGB.

4. The method of claim 1 wherein the association is executed by applying a flare color trained classifier.

5. The method of claim 4 wherein the flare color trained classifier is trained as a function of data representing flare constituent.

6. The method of claim 1 wherein the region of interest is identified by a minimum bounding rectangle.

7. The method of claim 1 and further comprising trend monitoring of the flare with respect to time and vented gas flow data obtained from flow sensors.

8. A computer readable medium having instructions for causing a computer to implement a method of identifying characteristics of a flare, the method comprising:

obtaining a digital image of the flare;

identifying a region of interest of said captured image;

identifying flare pixels within said region of interest;

extracting color information from the flare pixels;

associating the extracted color information to characterizations of said flare; and

providing an operator alert when such characterizations indicate an abnormal flare.

9. The computer readable medium of claim 8 wherein identification of flare pixels is executed by computer applying a classifier trained for flare detection.

10. The computer readable medium of claim 8 wherein color information is extracted in a color space selected from the group consisting of RGB, HSV, YCbCr, and Normalized RGB.

11. The computer readable medium of claim 8 wherein the association is executed by applying a flare color trained classifier.

12. The computer readable medium of claim 8 wherein the flare color trained classifier is trained as a function of data representing flare constituent.

13. A computer system that identifies characteristics of a flare, the system comprising:

a memory that stores a digital image of the flare;

means for identifying a region of interest of said captured image;

means for identifying flare pixels within said region of interest;

means for extracting color information from the flare pixels;

a classifier that associates the extracted color information to characterizations of said flare to identify an abnormal flare.

14. The computer system of claim 13 and further comprising a color video camera coupled to the memory that provides the digital image of the flare.

15. The computer system of claim 14 wherein the color video camera provides a digital video stream of the flare to the memory.

16. The computer system of claim 14 wherein the camera is positioned and set such that the flare occupies at least 10% of a camera field of view.

17. The computer system of claim 13 wherein the classifier provides an alert when characterizations represent an abnormal flare for a predetermined period of time.

18. A computer implemented method that identifies characteristics of a flare, the method comprising:

determining a broad region of interest on a video image where the flare is likely to be present;

detecting a flare area;

analyzing color constituents of the flare to detect abnormal flare conditions; and

providing an alarm when abnormal flare conditions are detected.

19. The method of claim 18 wherein the flare area is detected using a first manually trained classifier, and wherein color constituents of the flare are detected using a second manually trained classifier.