Method of isomorphic singular manifold projection and still/video imagery compression

• US 6,167,155 A
• Filed: 07/28/1997
• Issued: 12/26/2000
• Est. Priority Date: 07/28/1997
• Status: Expired due to Term

• Current Assignee: United California Bank (Los Angeles, California) (Old) (Bank of Montreal)
• Original Assignee: Physical Optics Corporation
• Inventors: Kostrzewski, Andrew, Jannson, Tomasz P., Ternovskiy, Igor
• Primary Examiner(s): Boudreau, Leo H.
• Assistant Examiner(s): SHERALI, ISHRAT I
• Application Number:
  US08/901,832
• Time in Patent Office: 1,247 Days
• Field of Search: 382/236-239, 382/250, 382/232, 348/616, 348/700, 348/699, 348/390-394, 348/400-421, 386/111, 360/8
• US Class Current: 382/232
• CPC Class Codes: G06T 9/001 Model-based coding, e.g. wire frame

Abstract

Methods and apparatuses for still image compression, video compression and automatic target recognition are disclosed. The method of still image compression uses isomorphic singular manifold projection whereby surfaces of objects having singular manifold representations are represented by best match canonical polynomials to arrive at a model representation. The model representation is compared with the original representation to arrive at a difference. If the difference exceeds a predetermined threshold, the difference data are saved and compressed using standard lossy compression. The coefficients from the best match polynomial together with the difference data, if any, are then compressed using lossless compression. The method of motion estimation for enhanced video compression sends I frames on an "as-needed" basis, based on comparing the error between segments of a current frame and a predicted frame. If the error exceeds a predetermined threshold, which can be based on program content, the next frame sent will be an I frame. The method of automatic target recognition (ATR) including tracking, zooming, and image enhancement, uses isomorphic singular manifold projection to separate texture and sculpture portions of an image. Soft ATR is then used on the sculptured portion and hard ATR is used on the texture portion.

Claims

1. A method of compressing video, said method comprising the steps of:

determining an actual error between corresponding blocks of a current frame and a predicted frame corresponding to the current frame;

repeating said determining step for a successive pair of the corresponding blocks;

accumulating the actual errors to generate an accumulated error;

inserting an I-frame as the next subsequent frame after the current frame if said accumulated error exceeds a predetermined threshold; and

repeating said first repeating, accumulating and inserting steps until said accumulated error exceeds said threshold.

2. A method of compressing video, said method comprising the steps of:

taking a frame F₀ of the video;

predicting a subsequent frame based on said frame F₀ and segmenting the predicted frame into search blocks;

taking an input frame F₁ of the video, and segmenting the input frame F₁ into search blocks corresponding to the search blocks of said predicted frame;

determining an actual error between corresponding search blocks of the input frame F₁ and the predicted frame;

accumulating the actual errors of different ones of the corresponding search blocks on a block-by-block basis to generate an accumulated error; and

comparing, after each accumulation of the errors, the accumulated error to a threshold and inserting an I-frame as the next subsequent frame if the accumulated error exceeds said threshold.

3. The invention as defined in claim 2, further comprising the steps of segmenting the frame F₀ into microblocks within said search blocks and normalizing the error by dividing by the number of microblocks.

4. The method as defined in claim 2, wherein said threshold is based upon video content.

5. The method as defined in claim 2, wherein said I frame is inserted in place of said frame F₀ if said error exceeds said threshold.

6. The method as defined in claim 2, further comprising the step of dynamically changing the compression ratio on a frame by frame basis based upon said error.

7. The invention as defined in claim 6, wherein one of a genetic algorithm, neural network, and fuzzy logic are used to determine the necessary change in compression ratio.

8. A method of compressed video transmission, the method comprising the steps of:

taking a first frame F₀ ;

segmenting said frame F₀ into blocks;

compressing each said block by representing it with a canonical polynomial having original coefficients;

predicting a frame P corresponding to said frame F₀ ;

determining an error between each block in said frame F₀ and a corresponding block of said predicted frame P;

accumulating the error determined between each block in said frame F₀ and said corresponding block of predicted frame P; and

comparing, after accumulating each error, the accumulated error to a threshold and if the accumulated error exceeds said threshold, inserting an I-frame as the next subsequent frame to said frame F₀.

9. A method as defined in claim 8 wherein said I frame was previously compressed by representing it with canonical polynomials.

10. A method of compressed video transmission, the method comprising the steps of:

taking a first frame F₀ ;

segmenting said frame F₀ into blocks;

compressing each said block by representing it with a canonical polynomial having original coefficients;

predicting a frame P corresponding to said frame F₀ based on a frame prior to F₀ ;

comparing each block in said frame F₀ with said predicted frame P to determine if there is a match;

if a match is found in said frame P for a block in said frame F₀, sending said coefficients of said polynomial for that block to a decoder;

if a match is not found for a block in said frame F₀, generating new coefficients of said polynomial representing said predicted frame P and sending said new coefficients to said decoder;

reconstructing said frame F₀ in said decoder from said original coefficients and said new coefficients sent to said decoder;

calculating an error between said frame F₀ and said predicted frame P based on said comparing each block in said frame F₀ with a corresponding block of said predicted frame P;

comparing said error to a threshold;

if said error does not exceed said threshold, sending a B or P frame as the next subsequent frame to said frame F₀ ;

if said error exceeds said threshold, inserting an I-frame as the next subsequent frame to said frame F₀.

11. The method as defined in claim 10, further comprising the steps of replacing said original coefficients, and said new coefficients with an I frame.

12. A method of compressing a video image, said method comprising the steps of:

taking a frame F₀ ;

segmenting F₀ into blocks and defining motion vectors for said blocks to predict a subsequent frame P comprising a plurality of blocks;

taking a subsequent frame F₁ of the video corresponding to said predicted frame P and segmenting F₁ into blocks corresponding to said blocks of said predicted frame;

defining an actual error between each said block in said frame F₁ and said corresponding block in said frame P;

accumulating the actual errors one error at a time to generate an accumulated error;

based on said accumulated error, sending an I-frame as the next subsequent frame to said frame F₀ ; and

repeating said defining, accumulating and sending steps for each corresponding pair of blocks of said frame F₁ and said frame P.

13. A method of video compression utilizing motion estimation comprising the steps of:

(a) comparing a segment of the current frame to be encoded to a corresponding segment of the previous frame to arrive at an error difference;

(b) repeating the calculation of Step (a) until error differences have been determined for all segments of the frame;

(c) summing said error differences to arrive at a cumulative error difference for that frame;

(d) comparing said cumulative error difference to a threshold and inserting a new I frame if said cumulative error exceeds said threshold.

14. The method of claim 13 further comprising adjusting the compression ratio on a frame by frame basis based on feedback and further using such feedback to change parameters of the compression engine.

15. The method of claim 13 further comprising lowering bandwidth by increasing the compression ratio based on feedback indicating that there is a high bit error rate.

16. The method of claim 14 wherein the change in compression ratio and bandwidth are determined by one of, a genetic algorithm, a neural network and fuzzy logic.

17. The method of claim 13 further comprising the step of presetting said threshold based upon the program content or type of video images to be compressed such as action, documentary, or nature.

18. A method of video compression, said method comprising the steps of:

deriving a plurality of errors between microblocks in a current frame of video and corresponding microblocks in a predicted frame;

accumulating said errors to generate an accumulated error;

comparing said accumulated error to a threshold after each said error is added to said accumulated error; and

sending an I frame when said threshold is exceeded.

19. The method of claim 18 further comprising sending the I frame as the current frame so as to replace error data.

20. The method of claim 18 wherein the threshold is varied in accordance with statistical information of the video based on program content such as the number of cuts, the amount of action, etc.

21. The method of claim 18 further comprising varying said threshold by using genetic algorithms.

22. The method of claim 18 further comprising varying said threshold by using fuzzy logic.

23. A method of video compression, said method comprising the steps of:

deriving a plurality of errors between microblocks in a current frame of video and corresponding microblocks in a predicted frame;

accumulating said errors to generate an accumulated error;

comparing said accumulated error to a threshold after each said error is added to said accumulated error;

sending an I frame when said threshold is exceeded; and

applying spatial hypercompression utilizing isomorphic singular manifold representations to each I frame inserted into the video bitstream.

24. A method of compressing video, said method comprising the steps of:

(a) calculating the actual error between a first block of a current frame and a corresponding block of a predicted frame;

(b) storing the calculated actual error as a first block actual error;

(c) calculating the actual error between a second block of the current frame and a corresponding block of the predicted frame to create a second block actual error;

(d) adding the second block actual error to the stored first block actual error to create an accumulated actual block error;

(e) calculating the actual error between a subsequent block of the current frame and a corresponding block of the predicted frame to create a subsequent block error;

(f) adding the subsequent block actual error to the accumulated actual block error to create a new accumulated actual block error;

(g) repeating steps (e) and (f) with respect to each additional subsequent block of the current frame, one at a time, until the accumulated actual error exceeds a predetermined threshold; and

(h) inserting an I-frame if the accumulated actual block error or the new accumulated actual block error exceeds the threshold.