Methods of representing and analysing images
A method of representing at least one image comprises deriving at least one descriptor based on color information and color interrelation information for at least one region of the image, the descriptor having at least one descriptor element, derived using values of pixels in said region, wherein at least one descriptor element for a region is derived using a non-wavelet transform. The representations may be used for image comparisons.
1. A method of representing an image, using one or more processors, comprising: deriving, using a processor, at least one descriptor based on colour information and colour interrelation information for at least one region of the image, the descriptor being derived using values of pixels in said region and having a plurality of descriptor elements, wherein at least one descriptor element for a region is derived using only a subset of pixels in said region, at least one descriptor element captures colour information, and at least one descriptor element captures colour interrelation information, and wherein deriving a plurality of different descriptor elements comprises using the same calculations for different numbers of pixels, or different sized regions, that is, at different scales.
2. The method of claim 1 wherein at least one descriptor element represents a pixel average for said region, and at least one descriptor element represents the difference function involving pixels in said region.
3. The method of claim 2 wherein at least one descriptor element represents a difference between two pixel values, or a difference between two sums of pixel values.
4. The method of claim 1 wherein descriptors are produced for a plurality of over-lapping or non-overlapping regions of an image, such as blocks of size n×n pixels.
5. The method of claim 1 further comprising processing the image to produce a plurality of representations of the image at a plurality of scales by downsampling or averaging, and deriving descriptors for said plurality of scales, or processing the image for each of a plurality of colour channels and deriving descriptors for each of said channels.
6. The method of claim 1 wherein different processes for deriving descriptors are used for different regions or different scales or different colour channels.
7. The method of claim 1 comprising associating a plurality of descriptors for an image to form a single descriptor.
8. The method of claim 1 wherein all the descriptor elements are derived for a single colour channel.
9. The method of claim 1 using colour space YCrCb.
10. The method of claim 1 , further comprising deriving descriptor elements for luminance channel Y.
11. A method of representing a group of images using one or more processors, the method comprising:
deriving, using a processor, a representation of each image by deriving at least one descriptor based on colour information and colour interrelation information for at least one region of the image, the descriptor being derived using values of pixels in said region and having a plurality of descriptor elements, wherein at least one descriptor element for a region is derived using only a subset of pixels in said region, at least one descriptor element captures colour information, and at least one descriptor element captures colour interrelation information, and wherein deriving a plurality of different descriptor elements includes using the same calculations for different numbers of pixels, or different sized regions, that is, at different scales; and
using said representations to represent said group of images by deriving a function of said representations to represent the group of images, wherein said function includes a statistical measure selected from an average, a median, a mean, or a covariance.
12. A method of comparing images or groups of images comprising:
comparing representations derived using the method of claim 11 to determine similarity or for image retrieval, for example,
comprising comparing images which are part of a sequence of images and wherein the result of the comparison is used to determine whether a frame discontinuity or event, such as a scene change or lighting effect, has occurred between the images.
13. An apparatus for comparing images or groups of images according to the method of claim 12 comprising storage means for storing image data or image descriptors, and processing means for executing said method.
14. A method of encoding an image or group of images using the method of claim 1 or 11 , use of data encoded according to such a method by means of transmission, relay or reception, or a method of decoding an image or sequence of images encoded using such a method.
15. An apparatus, for representing an image or group of images according to the method of claim 1 or 11 comprising storage means for storing image data or image descriptors, and processing means for executing said method.