Monitoring an output from a camera

1. A method of monitoring output data from a video camera, comprising the steps of:

analyzing input data to produce a model of a background image;

comparing incoming image data against said model of a background image to determine pixels within incoming images that define a foreground object;

for each incoming image, deriving a level of activity by measuring the proportion of pixels in the image that belong to a foreground object;

comparing the level of activity of each incoming image with a threshold value to identify images having a high level of activity created by the introduction of a foreground object;

identifying periods of activity, each said period of activity being composed of a plurality of consecutive images having a high level of activity; and

generating an exemplar image from each said period of activity,

such that said step of generating said exemplar image comprises for each period of activity

(i) selecting an exemplar image from a peak level of activity occurring within the period of activity;

(ii) selecting an exemplar image from substantially the center of the period of activity, or

(iii) combining a plurality of images from the period of activity.

2. A method according to claim 1 , wherein said step of generating said exemplar image comprises combining a plurality of images from the period of activity, and said exemplar image is generated by combining a weighted average of said plurality of images from said period of activity.

3. A method according to claim 1 , wherein each said exemplar image is cropped so as to remove the background image and retain an image of the foreground activity.

4. Apparatus for displaying graphical output to an operator comprising:

a plurality of detection devices for monitoring a region;

a processor configured to analyze data from said detection devices to identify periods of activity; and

display means arranged to display an exemplar image from said periods of activity,

wherein said processor is configured to perform the steps of:

analyzing input data to produce a model of a background image;

comparing incoming image data against said model of a background image to determine pixels within incoming images that define a foreground object;

for each incoming image, deriving a level of activity by measuring the proportion of pixels in the image that belong to a foreground object;

comparing the level of activity of each incoming image with a threshold value to identifying images having a high level of activity created by the introduction of a foreground object;

identifying periods of activity, each of said periods of activity composed of a plurality of consecutive images having a high level of activity; and

generating an exemplar image from each of said periods of activity, such that said step of generating said exemplar image comprises, for each period of activity,

(i) selecting an exemplar image from an activity peak occurring within the period of activity, or

(ii) selecting an exemplar image from substantially the center of said period of activity.

5. Apparatus according to claim 4 , wherein said display means is configured to display a plurality of said exemplar images from a plurality of periods of activity over a timeline.

6. Apparatus according to claim 4 , wherein said processor is configured to create a report after a predetermined period, and at least one of said exemplar images is included in said report.

7. Apparatus according to claim 6 , wherein all of said exemplar images are included in said report.

8. Apparatus according to claim 6 , wherein said report is displayed on said display means.

9. Apparatus according to claim 6 , further comprising a printing device configured to print said report.

10. Apparatus according to claim 6 , further comprising an input device configured to allow an operator to manually add text to said report.

11. Apparatus according to claim 6 , further comprising a network configured to allow other users to access said report.

12. A computer-readable medium having computer-readable instructions executable by a computer or by a network of computers such that when executing said instructions and computer(s) will perform the steps of:

analyzing input data to produce a model of a background image;

comparing incoming image data against said model of a background image to determine pixels within incoming images that define a foreground object;

for each incoming image, deriving a level of activity by measuring the proportion of pixels in the image that belong to a foreground object;

comparing the level of activity of each incoming image with a threshold value to identify images having a high level of activity created by the introduction of a foreground object;

identifying periods of activity, each of said periods of activity composed of a plurality of consecutive images having a high level of activity; and

generating an exemplar image from each of said periods of activity,

such that said step of generating an exemplar image comprises, for each period of activity,

(i) selecting an exemplar image from an activity peak occurring within the period of activity,

(ii) selecting an exemplar image from substantially the center of the period of activity, or

(iii) combining a plurality of images from the period of activity.

13. A computer-readable medium having computer-readable instructions according to claim 12 , such that said step of generating said exemplar image comprises combining a plurality of images from the period of activity, and when executing said instructions a computer will generate said exemplar image by combining a weighted average of said plurality of images from said period of activity.

14. A computer-readable medium having computer-readable instructions according to claim 12 , such that when executing said instructions a computer will crop each said exemplar image so as to remove the background image and retain an image of the foreground activity.