Method and system for automated video image focus change detection and classification

1. A system for evaluating a focus status of a video camera, the system comprising:

a processor; and a memory storing instructions that, when executed by the processor, cause the processor to perform operations including:

receiving, at a receiver, a video image from a video server in accordance with a predetermined schedule;

determining a focus classification procedure to perform on the received video image; classifying the focus status of the video image in accordance with the focus classification procedure and determining, from the video image whether the video camera is in a state of focus; and

clearing a warning count when the video image is determined to be in a state of focus and setting an alarm indication when the warning count exceeds a predetermined threshold wherein the focus classification procedure comprises:

incrementally building an estimate of a background of a scene;

storing a reference image based on the estimate;

retrieving the reference image;

segmenting each of the video image and the reference image into a respective array of sub-images;

calculating a focus metric for each sub-image of the video image and the reference image;

comparing focus metrics between the sub-images of the video image and the sub-images of the reference image;

calculating a percent of sub-images that have a focus metric difference greater than a first parameter; and

determining whether the video camera is in the state of focus based on whether a percent of the image that is out of focus is greater than a second parameter;

wherein the focus metric may change for an out of focus image based on state of focus.

2. The system according to claim 1 , wherein the operations further include,

incrementing the warning count a first time when the video image is determined to be in a state of non-focus as a result of a first determination, and setting a warning status when the warning count is incremented.

3. The system according to claim 2 , wherein the operations further include,

clearing the incremented count when the video image is determined to be in the state of focus as a result of a second determination.

4. The system according to claim 2 , wherein the operations further include,

incrementing the warning count a second time when the video image is determined to be in the state of focus as a result of a second determination; and

setting the alarm indication when the warning count, incremented the second time, exceeds the predetermined threshold of at least two.

5. The system according to claim 1 , wherein the operations further include,

selecting one or more distinct focus classification procedures; and

classifying the focus status of an image of the video camera in accordance with the selected focus classification procedure.

6. The system according to claim 1 , wherein the operations further include:

selecting the focus classification procedure from a plurality of focus classification procedures; and

a second focus classification procedure of the plurality of focus classification procedures comprises:

segmenting the video image into an array of sub-images;

calculating at least one metric associated with a focus determination for each sub-image;

applying a blur to the video image to create a blurred image approximating what the video image would be if the video camera were in an out of focus state;

segmenting the blurred image in order to estimate how the at least one metric may change for an out of focused image;

storing the video image and the blurred image;

predicting an amount of blur for each sub-image;

comparing the predicted amount of blur to an acceptable amount of blur in the video image;

calculating a percent of sub-images that have a predicted blur greater than or equal to a first parameter, the percent representing a confidence that a whole of the video image is out of focus; and

determining whether the video camera is in the state of focus based on whether the percent of the video image considered out of focus is greater than a second parameter.

7. A method for evaluating a focus status of a video camera, the method comprising:

receiving, at a receiver a video image from a video server in accordance with a predetermined schedule;

determining a focus classification procedure to perform on the received video image;

classifying the focus status of the video image in accordance with the focus classification procedure and determining, from the video image, whether the video camera is in a state of focus; and

clearing a warning count when the video image is determined to be in a state of focus and setting an alarm indication when the warning count exceeds a predetermined threshold{circumflex over ( )} wherein the focus classification procedure comprises:

incrementally building an estimate of a background of a scene;

storing a reference image based on the estimate;

retrieving the reference image;

segmenting each of the video image and the reference image into a respective array of sub-images;

calculating a focus metric for each sub-image of the video image and the reference image;

comparing focus metrics between the sub-images of the video image and the sub-images of the reference image;

calculating a percent of sub-images that have a focus metric difference of greater than a first parameter; and

determining whether the video camera is in the state of focus based on whether a percent of the image that is out of focus is greater than a second parameter;

wherein the focus metric may change for an out of focus image based on state of focus.

8. The method according to claim 7 , wherein the operations further include,

incrementing the warning count a first time when the video image is determined to be in a state of non-focus as a result of a first determination, and setting a warning status when the warning count is incremented.

9. The method according to claim 8 , wherein the operations further include,

clearing the incremented count when the video image is determined to be in the state of focus as a result of a second determination.

10. The method according to claim 8 , wherein the operations further include,

incrementing the warning count a second time when the video image is determined to be in the state of focus as a result of a second determination; and

setting the alarm indication when the warning count, incremented the second time, exceeds the predetermined threshold of at least two.

11. The method according to claim 7 , wherein the operations further include,

selecting one or more of focus distinct classification; and

classifying the focus status of an image of the video camera in accordance with the selected focus classification procedure.

12. The method according to claim 7 , wherein the operations further include:

selecting the focus classification procedure from a plurality of focus classification procedures; and

a second focus classification procedure of the plurality of focus classification procedures comprises:

segmenting the video image into an array of sub-images;

calculating at least one metric associated with a focus determination for each sub-image;

applying a blur to the video image to create a blurred image approximating what the video image would be if the video camera were in an out of focus state;

segmenting the blurred image in order to estimate how the at least one metric may change for an out of focused image;

storing the video image and the blurred image;

predicting an amount of blur for each sub-image;

comparing the predicted amount of blur to an acceptable amount of blur in the video image;

calculating a percent of sub-images that have a predicted blur greater than or equal to a first parameter, the percent representing a confidence that a whole of the video image is out of focus; and

determining whether the video camera is in the state of focus based on whether the percent of the video image considered out of focus is greater than a second parameter.

13. A non-transitory computer readable storage medium that stores a set of executable instructions for evaluating a focus status of a video camera, the executable instructions, when executed by a processor, causing operations to be performed comprising:

receiving, at a receiver a video image from a video server in accordance with a predetermined schedule;

determining a focus classification procedure to perform on the received video image;

classifying the focus status of the video image in accordance with the focus classification procedure and determining, from the video image, whether the video camera is in a state of focus; and

clearing a warning count when the video image is determined to be in a state of focus and setting an alarm indication when the warning count exceeds a predetermined threshold wherein the focus classification procedure comprises:

incrementally building an estimate of a background of a scene;

storing a reference image based on the estimate;

retrieving the reference image;

segmenting each of the video image and the reference image into a respective array of sub-images;

calculating a focus metric for each sub-image of the video image and the reference image;

comparing focus metrics between the sub-images of the video image and the sub-images of the reference image;

calculating a percent of sub-images that have a focus metric difference of greater than a first parameter; and

determining whether the video camera is in the state of focus based on whether a percent of the image that is out of focus is greater than a second parameter;

wherein the focus metric may change for an out of focus image based on state of focus.

14. The non-transitory computer readable storage medium according to claim 13 , wherein the operations further include,

incrementing the warning count a first time when the video image is determined to be in a state of non-focus as a result of a first determination, and setting a warning status when the warning count is incremented.

15. The non-transitory computer readable storage medium according to claim 14 , wherein the operations further include,

clearing the incremented count when the video image is determined to be in the state of focus as a result of a second determination.

16. The non-transitory computer readable storage medium according to claim 14 , wherein the operations further include,

incrementing the warning count a second time when the video image is determined to be in the state of focus as a result of a second determination; and

setting the alarm indication when the warning count, incremented the second time, exceeds the predetermined threshold of at least two.

17. The non-transitory computer readable storage medium according to claim 13 , wherein the operations further include:

selecting the focus classification procedure from a plurality of focus classification procedures; and

a second focus classification procedure of the plurality of focus classification procedures comprises:

segmenting the video image into an array of sub-images;

calculating at least one metric associated with a focus determination for each sub-image;

applying a blur to the video image to create a blurred image approximating what the video image would be if the video camera were in an out of focus state;

segmenting the blurred image in order to estimate how the at least one metric may change for an out of focused image;

storing the video image and the blurred image;

predicting an amount of blur for each sub-image;

comparing the predicted amount of blur to an acceptable amount of blur in the video image;

calculating a percent of sub-images that have a predicted blur greater than or equal to a first parameter, the percent representing a confidence that a whole of the video image is out of focus; and

determining whether the video camera is in the state of focus based on whether the percent of the video image considered out of focus is greater than a second parameter.