Quantifying gas leak rates using frame images acquired by a camera

1. A method of quantifying gas leak rate, comprising:

receiving a first plurality of image frames acquired with a first camera and comprising a plume from a gas leak source;

identifying a plurality of pixels corresponding to the plume in a first image frame of the first plurality of images frames;

calculating a plurality of gas concentration path lengths of the plume for the plurality of pixels in the first image frame;

calculating, based on the first image frame and a second image frame, an image velocity field of the plume comprising a plurality of displacement vectors for the plurality of pixels;

identifying, within the first image, a plurality of pixels corresponding to the gas leak source and generating a closed boundary enclosing an entirety of the plurality of pixels corresponding to the gas leak source;

calculating a first gas leak rate in the first image frame by calculating a volume rate of the plume flowing across the closed boundary based on:

the image velocity field;

the plurality of gas concentration path lengths; and

a time interval between the first image frame and the second image frame,

wherein the method further comprises:

post-processing the first plurality of image frames prior to calculating of the image velocity field of the plume;

calculating the gas leak rate using the first gas leak rate, and

wherein the first plurality of post-processed image frames are concentration path length map images.

2. The method of claim 1 , further comprising:

calculating an average distance of the plurality of displacement vectors;

applying a sub-pixel flow correction method in response to the average distance being less than 1 pixel.

3. The method of claim 2 , wherein the sub-pixel flow correction method comprises:

scaling the calculated first gas leak rate by a factor of the average distance of the image velocity field at each of the plurality of pixels.

4. The method of claim 1 , further comprising:

calculating a second gas leak rate in the second image frame; and

calculating an average gas leak rate based on the first gas leak rate and the second gas leak rate.

5. The method of claim 1 , wherein calculating the volume rate comprises:

identifying, within the closed boundary, a first contributing pixel comprising a displacement vector that points from the first contributing pixel to outside the closed boundary;

identifying, within the closed boundary, a second contributing pixel comprising a second displacement vector that points from the second contributing pixel to outside the closed boundary;

calculating a first gas volume from the first contributing pixel by multiplying the concentration path length of the first contributing pixel by the real-world size of the first contributing pixel;

calculating a second gas volume from the second contributing pixel by multiplying the concentration path length of the second contributing pixel by the real-world size of the second contributing pixel; and

generating a total gas volume by summing the first gas volume and the second gas volume; and

dividing the total gas volume by the time interval between the first image frame and the second image frame.

6. The method of claim 1 , wherein the image velocity field of the plume is calculated by an optical flow method.

7. The method of claim 1 , wherein the image velocity field of the plume is calculated by a geometric flow method.

8. The method of claim 1 , further comprising:

obtaining a second plurality of image frames acquired by a second camera and comprising the plume from the gas leak source,

wherein the first camera and the second camera are at different viewing angles;

calculating a second gas leak rate using the second plurality of image frames; and

averaging the first gas leak rate and the second gas leak rate.

9. The method of claim 8 , wherein the viewing angle of the first camera is orthogonal to the viewing angle of the second camera.

10. A non-transitory computer readable medium (CRM) storing computer readable program code embodied therein that:

receives a first plurality of image frames acquired with a first camera and comprising a plume from a gas leak source;

identifies a plurality of pixels corresponding to the plume in a first image frame of the first plurality of images frames;

calculates a plurality of gas concentration path lengths of the plume for the plurality of pixels in the first image frame;

calculates, based on the first image frame and a second image frame, an image velocity field of the plume comprising a plurality of displacement vectors for the plurality of pixels;

identifies, within the first image, a plurality of pixels corresponding to the gas leak source and generating a closed boundary enclosing an entirety of the plurality of pixels corresponding to the gas leak source;

calculates a first gas leak rate in the first image frame by calculating a volume rate of the plume flowing across the closed boundary based on:

the image velocity field;

the plurality of gas concentration path lengths; and

a time interval between the first image frame and the second image frame;

post-processes the first plurality of image frames prior to calculating of the image velocity field of the plume;

calculates the gas leak rate using the first gas leak rate, and

wherein the first plurality of post-processed image frames are concentration path length map images.

11. The non-transitory computer readable medium (CRM) of claim 10 storing computer readable program code embodied therein that further:

calculates an average distance of the plurality of displacement vectors;

applies a sub-pixel flow correction method in response to the average distance being less than 1 pixel.

12. The non-transitory computer readable medium (CRM) of claim 10 storing computer readable program code embodied therein that further:

calculates a second gas leak rate in the second image frame; and

calculates an average gas leak rate based on the first gas leak rate and the second gas leak rate.

13. The non-transitory computer readable medium (CRM) of claim 10 storing computer readable program code embodied therein, wherein calculating the volume rate comprises:

identifying, within the closed boundary, a first contributing pixel comprising a displacement vector that points from the first contributing pixel to outside the closed boundary;

identifying, within the closed boundary, a second contributing pixel comprising a second displacement vector that points from the second contributing pixel to outside the closed boundary;

calculating a first gas volume from the first contributing pixel by multiplying the concentration path length of the first contributing pixel by the real-world size of the first contributing pixel;

calculating a second gas volume from the second contributing pixel by multiplying the concentration path length of the second contributing pixel by the real-world size of the second contributing pixel; and

generating a total gas volume by summing the first gas volume and the second gas volume; and

dividing the total gas volume by the time interval between the first image frame and the second image frame.

14. A system for quantifying gas leak rate, comprising:

a memory; and

a processor that:

receives a first plurality of image frames acquired with a first camera and comprising a plume from a gas leak source;

identifies a plurality of pixels corresponding to the plume in a first image frame of the first plurality of images frames;

calculates a plurality of gas concentration path lengths of the plume for the plurality of pixels in the first image frame;

calculates, based on the first image frame and a second image frame, an image velocity field of the plume comprising a plurality of displacement vectors for the plurality of pixels;

identifies, within the first image, a plurality of pixels corresponding to the gas leak source and generating a closed boundary enclosing an entirety of the plurality of pixels corresponding to the gas leak source;

calculates a first gas leak rate in the first image frame by calculating a volume rate of the plume flowing across the closed boundary based on:

the image velocity field;

the plurality of gas concentration path lengths; and

a time interval between the first image frame and the second image frame,

wherein the processor further:

post-processes the first plurality of image frames prior to calculating of the image velocity field of the plume;

calculates the gas leak rate using the first gas leak rate, and

wherein the first plurality of post-processed image frames are concentration path length map images.

15. The system of claim 14 , wherein the processor further:

calculates an average distance of the plurality of displacement vectors;

applies a sub-pixel flow correction method in response to the average distance being less than 1 pixel.

16. The system of claim 14 , wherein the processor further:

calculates a second gas leak rate in the second image frame; and

calculates an average gas leak rate based on the first gas leak rate and the second gas leak rate.

17. The system of claim 14 , wherein calculating the volume rate comprises:

identifying, within the closed boundary, a first contributing pixel comprising a displacement vector that points from the first contributing pixel to outside the closed boundary;

identifying, within the closed boundary, a second contributing pixel comprising a second displacement vector that points from the second contributing pixel to outside the closed boundary;

calculating a first gas volume from the first contributing pixel by multiplying the concentration path length of the first contributing pixel by the real-world size of the first contributing pixel;

calculating a second gas volume from the second contributing pixel by multiplying the concentration path length of the second contributing pixel by the real-world size of the second contributing pixel; and

generating a total gas volume by summing the first gas volume and the second gas volume; and

dividing the total gas volume by the time interval between the first image frame and the second image frame.