Video image mapping system

• US 5,173,948 A
• Filed: 03/29/1991
• Issued: 12/22/1992
• Est. Priority Date: 03/29/1991
• Status: Expired due to Term

• Current Assignee: Grass Valley USA LLC (Vislink Technologies, Inc.)
• Original Assignee: The Grass Valley Group, Inc. (Belden Inc.)
• Inventors: Lawrence, Frank S., Blackham, Raymond C.
• Primary Examiner(s): COUSO, JOSE L
• Application Number:
  US07/677,548
• Time in Patent Office: 634 Days
• Field of Search: 382/41, 382/44-46, 340/424, 340/427, 358/138, 358/160
• US Class Current: 382/293
• CPC Class Codes: G06T 15/04 Texture mapping
• CPC Class Codes: H04N 5/2628 Alteration of picture size, shape, position or orientation, e.g. zooming, rotation, rolling, perspective, translation

Abstract

A video mapping system includes a weighting coefficients generator and a two-dimensional resampling filter that efficiently utilize memory resources and operate fast enough to process a video image stream without introducing artifacts through temporal interpolation. This system can accommodate non-linear and many-to-one mappings, allowing a source two-dimensional video image to be folded over itself in complicated ways to produce an image in the target coordinate system. The only restrictions are that the mapping be piecewise continuous and single-valued. Anti-aliasing filtering is implicit in the approach.

Claims

1. A method for mapping a video image in a first two-dimensional plane into a video image in a second two-dimensional plane according to an arbitrary mapping function, the method comprising the steps of:

mapping addresses of the video image in the first two-dimensional plane into the second two-dimensional plane to produce quadrilateral addresses;

translating an origin to produce sub-pixel addresses, pixel addresses and bank-set selection signals from the quadrilateral addresses;

generating weighting coefficients indicative of an area of a quadrilateral that is within each of a plurality of banks, the weighting coefficients being generated from the bank selection signals and the sub-pixel addresses;

using the generated coefficients and the pixel addresses to produce video data and video addresses in the second two-dimensional plane; and

organizing the video data and video addresses in the second two-dimensional plane to produce the video image in the second two-dimensional plane.

2. A method according to claim 1 wherein the generating weighting coefficient step comprises:

determining which of a plurality of geometric cases pertains from a relationship between the sub-pixel addresses and bank-set selection signals;

ascertaining secondary triangle ratios from appropriately selected sub-pixel addresses based on the determined case;

calculating primary triangle areas ABO, BDO, DCO and CAO from the sub-pixel addresses;

multiplying the primary triangle areas by the secondary triangle ratios according to the determined case to obtain secondary triangle areas; and

combining primary triangle areas with secondary triangle areas according to the determined case to obtain coefficients indicative of the area of a quadrilateral that is within each of the plurality of banks.

3. A method according to claim 1 wherein the using step comprises the steps of:

multiplying video data by an associated generated coefficient to produce a weighted video data element;

identifying same pixels from the pixel addresses to produce a same pixel signal;

accumulating weighted video data elements associated with the same output pixel into a running total based on the same pixel signal;

determining when there is no additional data associated with an output pixel; and

outputting a video address in the second two-dimensional plane and associated video data from the running total when it is determined that there is no additional data associated with that output pixel.

4. A method according to claim 3 wherein the step of identifying same pixels comprises the steps of:

latching a previous address; and

comparing the previous address with a current address to produce a same pixel signal when the previous address and the current address are equal.

5. A method according to claim 3 wherein the accumulating step comprises the steps of:

summing weighted video data elements associated with the same pixel together into a line running total according to the same pixel signal;

storing the line running total, an associated pixel address, and a count representing how many quadrilaterals on the present line contributed to the line running total; and

recalling the line running total from a previous line when a line running total on a present line should include a line running total on a previous line in order to produce a complete line running total.

6. A method according to claim 5 wherein the step of recalling the line running total comprises the steps of:

reading out the stored line running total and associated pixel address and the count while at a location on the current line corresponding to a location of the stored line running total on the previous line;

ascertaining from the read out contents whether any of the associated pixel addresses from the previous line are the same as the present address on the current line;

presenting the line running total associated with the associated pixel address for summation when one of the associated pixel addresses is the same as the present address on the current line; and

counting down from the count to keep the location on the current line synchronized with the previous line so that corresponding locations can be known.

7. A method according to claim 6 wherein the ascertaining step comprises the steps of:

latching read out contents; and

comparing the address portion of the read out contents with the present address on the current line and producing an equivalence indication when they are equal.

8. A method according to claim 1 wherein pixel addresses include a side indication.

9. A method according to claim 3 wherein pixel addresses include a side indication.

10. A method according to claim 5 wherein pixel addresses include a side indication.

11. A system for mapping a video image in a first two-dimensional plane into a video image in a second two-dimensional plane according to an arbitrary mapping function, the system comprising:

means for mapping addresses of the video image in the first two-dimensional plane into the second two-dimensional plane to produce quadrilateral addresses;

means for translating an origin to produce sub-pixel addresses, pixel addresses, and bank-set selection signals from the quadrilateral addresses;

means for generating weighting coefficients from the bank-set selection signals and the sub-pixel addresses to produce coefficients indicative of an area of a quadrilateral that is within each of a plurality of banks;

means for using the generated coefficients and the pixel addresses, to produce video data and video addresses in the second two-dimensional plane; and

means for organizing the video data and video addresses in the second two-dimensional plane to produce the video image in the second two-dimensional plane.

12. A system according to claim 11 wherein the means for generating weighting coefficients comprises:

means for determining which of a plurality of geometric cases pertains from a relationship between the sub-pixel addresses and bank-set selection signals;

means for ascertaining secondary triangle ratios from appropriately selected sub-pixel addresses based on the determined case;

means for calculating primary triangle areas ABO, BDO, DCO and CAO from the sub-pixel addresses;

means for multiplying the primary triangle areas by the secondary triangle ratios according to the determined case to obtain secondary triangle areas; and

means for combining primary triangle areas with secondary triangle areas according to the determined case to obtain coefficients indicative of the area of a quadrilateral that is within each of the plurality of banks.

13. A system according to claim 11 wherein the means for using comprises:

means for multiplying video data by an associated generated coefficient to produce a weighted video data element;

means for identifying same pixels from the pixel addresses to produce a same pixel signal when a current pixel has the same address as a previous pixel;

means for accumulating weighted video data elements associated with the same output pixel into a running total based on the same pixel signal;

means for determining when there is no additional data associated with a pixel; and

means for outputting a video address in the second two-dimensional plane and associated video data from the running total when it is determined that there is no additional data associated with a pixel.

14. A system according to claim 13 wherein the means for identifying same pixels comprises:

means for latching a previous address; and

means for comparing the previous address with a current address to produce a same pixel signal when the previous address and the current address are equal.

15. A system according to claim 13 wherein the means for accumulating comprises:

means for summing weighted video data elements associated with the same pixel together into a line running total according to the same pixel signal;

means for storing the line running total, an associated pixel address, and a count representing how many quadrilaterals on the present line contributed to the line running total; and

means for recalling the line running total from a previous line when a line running total on a present line should include a line running total on a previous line in order to produce a complete line running total.

16. A system according to claim 15 wherein the means for recalling the line running total comprises:

means for reading out the stored line running total and associated pixel address and the count while at a location on the current line corresponding to a location of the stored line running total on the previous line;

means for ascertaining from the read out contents whether any of the associated pixel addresses from the previous line are the same as the present address on the current line;

means for presenting the line running total associated with the associated pixel address for summation when one of the associated pixel addresses is the same as the present address on the current line; and

means for counting down from the count to keep the location on the current line synchronized with the previous line so that corresponding locations can be known.

17. A system according to claim 16 wherein the means for ascertaining comprises:

means for latching read out contents; and

means for comparing the address portion of the read out contents with the present address on the current line and producing an equivalence indication when they are equal.

18. A system according to claim 11 wherein pixel addresses include a side indication.

19. A system according to claim 13 wherein pixel addresses include a side indication.

20. A system according to claim 15 wherein pixel addresses include a side indication.