Apparatus and method for encoding data

• US 5,999,217 A
• Filed: 03/11/1997
• Issued: 12/07/1999
• Est. Priority Date: 06/06/1996
• Status: Expired due to Term

• Current Assignee: Charles Peter Berners-Lee
• Original Assignee: Charles Peter Berners-Lee
• Inventors: Berners-Lee, Charles Peter
• Primary Examiner(s): BRITTON, HOWARD W
• Application Number:
  US08/816,108
• Time in Patent Office: 1,001 Days
• Field of Search: 348/390, 348/384, 348/397, 348/398, 348/399, 348/411, 348/412, 348/413, 348/414
• US Class Current: 375/240.14
• CPC Class Codes: H04N 19/30 using hierarchical techniques, e.g. scalability H04N19/63 takes precedence
• CPC Class Codes: H04N 19/507 using conditional replenishment

Abstract

A method of encoding data, involves the steps of comparing an input image from a video camera with an image being built up, and stored in store. The images are processed by an analyser in accordance with a plurality of different algorithms stored in a store to produce a plurality of different data packets each representative of a partial change which can be effected to the image to be built up to bring it closer to the input image. The analyser compares the effects of the different data packets if implemented on the data image to be built up and provides each one with a score in accordance with a predetermined criteria. A comparator selects the packet with the best score under the predetermined criteria and uses it to alter the image in the store accordingly. The cycle is repeated to cause the image to be built up, to progressively converge on the input image.

Claims

1. Data encoding apparatus comprising

a first store for storing data representative of an input image,

a second store for storing data representative of an output image,

an analyser for comparing the data in the first and second stores, and conducting a plurality of different operations on the data, to produce a plurality of packets of data each representative of a different change to be applied to the output image, and to allocate a score to each packet indicative of the extent to which the packet will, if applied to the output image, improve the output image,

means for selecting the packet with the score representing the packet providing the most significant improvement, and

means for causing the selected packet to effect the change it represents to the data stored in the second store, whereby upon successive cycles to cause the data in the second store to progressively converge on the data stored in the first store.

2. Apparatus according to claim 1, including

an auxiliary second store for receiving said packets of data, and

a delay circuit for delaying the supply of said packets to the auxiliary store, the analyser being coupled to compare the data in the auxiliary second store with the data in the first store and if this comparison produces a better score than the comparison between the first and second stores triggering the auxiliary second store to alter the contents of the second store in accordance with the contents or selected portions of the contents of the auxiliary store.

3. Apparatus according to claim 1, wherein the analyser makes a pixel by pixel comparison of the images stored in the first and second stores.

4. Apparatus according to claim 1, including an algorithm store for storing a plurality of different algorithms whereby to enable the analyser to process the data defining the input and output images with the different algorithms.

5. Apparatus according to claim 4, wherein the analyser is instructed to process the data in the two stores with the different algorithms serially.

6. Apparatus according to claim 5, including

a current packet store for storing each packet produced by the analyser in turn,

a best packet store, and

a comparator for comparing the score of the data packet in the current packet store with the score of the data in the best packet store and transferring the packet in the current packet store to the best packet store when the score of the packet in the current packet store is better than the score of the packet in the best packet store.

7. Apparatus according to claim 4, wherein the analyser is instructed to process the data in the two stores with the different algorithms substantially simultaneously.

8. Apparatus according to claim 4, wherein the analyser is instructed to process the data in the two stores with different algorithms, some of the processes taking place simultaneously and others taking place serially.

9. Apparatus according to any preceding claim, including a selector for comparing the scores of the different packets produced in the different stores and selecting the packet with the best score for onward transmission to the second store.

10. A method of encoding data, comprising the steps of

comparing an input image with an image being built up,

processing the differences between the input image and the instantaneous built up image in accordance with one or more different algorithms to produce a plurality of different data packets each representative of a partial change which can be effected to the image to be built up to bring it closer to the input image,

comparing the effects of the different data packets if implemented on the data image to be built up and providing each one with a score indicative of significance of the said partial change,

selecting the packet with the best score and using it to alter the image to be built up accordingly, and

repeating the cycle to cause the image to be built up, to progressively converge on the input image.

11. A method according to claim 10, wherein the comparing step compares the two images pixel by pixel.

12. A method according to claim 10, wherein the processing step processes the different data packets substantially simultaneously.

13. A method according to claim 10, wherein the processing step processes the different data packets serially.

14. A method according to claim 10, wherein the processing step processes some of the different data packets serially and others simultaneously.

15. A method according to claim 10, wherein each data packet comprises the coordinates of a geometric shape.

16. A method according to claim 15, wherein the geometric shape is a rectangle.

17. A method of compressing data representative of an image, comprising the steps of

scanning an image in accordance with a selected one of a plurality of predetermined scanning regimes,

comparing the data produced as a result of the scan with data representing an image stored in a data store,

weighting the difference sensed, and in accordance with a predetermined criteria,

selecting one of said plurality of scanning regimes as determined by the weighting and producing difference data representative of at least some of the differences between the scanned image as scanned under the selected scanning regime and the image stored in the data store;

combining the difference data with data from the data store to produce data representative of a virtual image,

comparing the virtual image data with the data in the store and weighting the difference scanned in accordance with said predetermined algorithm to cause a scanning regime selected in accordance with the algorithm to scan the image,

producing difference data representative of at least some of the differences between the scanned image as scanned by the selected scanning regime and the image stored in the data store,

combining the difference data with data stored in the data store to produce a new virtual image,

repeating the steps of producing a new virtual image until a predetermined criteria determined at least in part by a comparison of the weightings is reached, and

transmitting the difference data used to produce a selected one of said virtual images to an output terminal and to said data store to update the data in the data store.