Image clustering method

1. A method of determining one or more event subsets within a plurality of images, each image being associated with time and location data specifying the time and location of capture of the image by an image capture device, said method comprising:

determining a time variable for each adjacent pair of images in a capture time ordered list of the plurality of images based on the time data associated with the images of the pair;

determining a distance variable for each adjacent pair of images in the ordered list of images based on the location data associated with the images of the pair;

generating speed data of the image capture at the time and location of capture of each image in the plurality of images; and

partitioning the ordered list of images into one or more event subsets on the basis of a cost function, said cost function being determined in accordance with a normalisation of the time variable and distance variable, wherein the time variable and the distance variable are weighted using the speed data.

2. A method according to claim 1 , wherein the step of determining speed data comprises at least one of:

(i) calculating the speed data from metadata associated with the images; and

(ii) receiving the speed data from a speed logging device.

3. A method according to claim 1 wherein the time variable is weighted more than the distance variable when the speed of the image capture device is over a predetermined threshold, and the distance variable is weighted more than the time variable when the speed of the image capture device is below the predetermined threshold.

4. A method according to claim 1 wherein the weighting comprises a selection between the time or distance.

5. A method according to claim 4 wherein the weighting is 90% distance and 10% time when the speed data is below a speed threshold and 100% time when the speed data is above the speed threshold.

6. A method according to claim 1 wherein the speed data is determined from the time variable and the distance variable.

7. A method according to claim 1 further comprising the step of ordering the plurality of images by a characteristic of the images to produce the capture time ordered list of images.

8. A method according to claim 7 wherein the characteristic of the image is at least one of time of capture, file name, and distance from a reference point.

9. A method according to claim 1 wherein the speed data is derived from at least one of periodic sampling of speed information or a GPS log of the location of the camera.

10. A method according to claim 1 wherein the cost function is a weighted average of the time variable and the distance variable.

11. A method according to claim 10 , wherein weights for the weighted average of the time variable and the distance variable are derived from the speed data.

12. A method according to claim 1 wherein the images are partitioned into the subsets based on a cost threshold such that where the cost function associated with an image is above the threshold, a new partitioned subset is created, and where the cost function is below the cost threshold, the image is assigned to an existing subset.

13. A method according to claim 1 wherein the time variable and the distance variable are derived from adjacent images in the ordered set of images.

14. A method according to claim 13 wherein the time variable comprises a time difference between the adjacent images, and the distance variable comprises a distance metric between the adjacent images.

15. A method according to claim 1 , wherein the speed data is derived from at least one of the speed of the photographer and the speed of the image capture device between the two images, and the speed of the image capture device when the image was taken.

16. A method according to claim 1 wherein the normalisation comprises one of mean-variance normalisation, feature warping, histogram equalisation or maximum a posteriori (MAP) adaptation.

17. A computer readable non-transitory storage medium having a computer program recorded thereon, the program being executable by computerised apparatus to perform the method of claim 1 .

18. An image processing apparatus for determining one or more event subsets within a plurality of images, each image being associated with time and location data specifying the time and location of capture of the image by an image capture device, the image processing apparatus comprising a processor and a memory associated with the processor, the memory having recorded thereon a program executable by the processor, the program comprising:

code for determining a time variable for each adjacent pair of images in a capture time ordered list of the plurality of images based on the time data associated with the images of the pair;

code for determining a distance variable for each adjacent pair of images in the ordered list of images based on the location data associated with the images of the pair;

code for generating speed data of the image capture at the time and location of capture of each image in the plurality of images; and

code for partitioning the ordered list of images into one or more event subsets on the basis of a cost function, said cost function being determined in accordance with a normalisation of the time variable and distance variable, wherein the time variable and the distance variable are weighted using the speed data.