Compression of a collection of images using pattern separation and re-organization

Patent metadata

Field	Value
Publication number	US-10038902-B2
Application number	US-61361309-A
Country	US
Kind code	B2
Filing date	Nov 6, 2009
Priority date	Nov 6, 2009
Publication date	Jul 31, 2018
Grant date	Jul 31, 2018

Abstract

Embodiments herein include an image manager that provides for image compression by rearranging an order of blocks from one or more images and then sorting and writing those blocks into one or more different images. This technique enables using a high-level of image compression to reduce a relatively large amount of pixels to a common subset of values than would ordinarily be possible with the original image(s). This can include extracting a plurality of blocks from a graphical digital image file. Each block from the graphical digital image file can be a group of pixels. The image manager analyzes each block to produce a corresponding variation value for each of the blocks, indicating a level of variation of pixel data within a respective block. The image manager sorts blocks according to the variation values, and can apply a level of image compression to each respective block, based on the variation value and/or average color of each respective block, to compress each respective block. For a given image or set of images, certain techniques disclosed herein can yield an image archive having a file size that is about two to four times smaller than conventional archiving techniques applied to the same given image or set of images.

First claim

The invention claimed is:

1. A computer-implemented method to perform preprocessing, compression, and archiving of a collection of digital image files comprising pixels, the method comprising:
with a processor, receiving a plurality of digital image files from a collection of digital image files;
in response to receiving the digital image files, extracting blocks, with the processor, from each of the digital image files, wherein each block comprises a subset of the pixels of the respective image file;
analyzing, with the processor, each of the blocks to determine a corresponding frequency of pixel data variation for each of the blocks;
creating, with the processor, a plurality of new digital image files each comprising a plurality of the blocks extracted from at least two of the received digital image files, wherein the blocks selected for each particular new digital image file are those of the blocks of a given range of frequencies of pixel data variation;
identifying, with the processor, a level of image compression to apply to each particular new digital image file such that the level of image compression is optimized for the frequencies of pixel data variation of the blocks in the particular new digital image file;
compressing, with the processor, each particular new digital image file using the level of image compression identified to apply to the particular new digital image file; and
with the processor, creating an image archive comprising the plurality of compressed new digital image files and storing the image archive to a memory.

2. The computer-implemented method of claim 1 , further comprising:
analyzing, with the processor, each of the blocks to produce a corresponding color value for each of the blocks, each corresponding color value indicating an average color of pixels within a respective block, wherein the blocks selected for each particular new digital image file are those of the blocks of a predetermined color range.

3. The computer-implemented method of claim 1 , wherein the identified level of image compression to apply to each particular new digital image file indicates a compression setting of a predetermined image compression algorithm; and
wherein compressing each particular new digital image file includes applying the predetermined image compression algorithm to the particular new digital image file according to the applicable identified level of image compression.

4. The computer-implemented method of claim 3 , wherein each particular new digital image file is compressed using a Joint Photographic Experts Group (JPEG) compression algorithm.

5. The computer-implemented method of claim 1 , wherein identifying a level of image compression to apply to each particular new image file includes:
for each particular new digital image file having small pixel data variation values in comparison to others of the new digital image files, identifying, with the processor, a low level of image compression to apply to the particular new digital image file, the low level of image compression resulting in a compressed new digital image file having relatively high image quality in comparison to others of the new digital image files; and
for each particular new digital image file having large pixel data variation values in comparison to others of the new digital image files, identifying, with the processor, a high level of image compression to apply to the particular new digital image file, the high level of image compression resulting in a compressed new digital image file having relatively low image quality in comparison to others of the new digital image files.

6. The computer-implemented method of claim 1 , further comprising:
with the processor, creating a storage map that identifies a location of each block in the received plurality of digital image files and storing the storage map in the memory;
with the processor, accessing the image archive from the memory and extracting the new digital image files from the image archive;
decompressing, with the processor, each of the new digital image files; and
with the processor, accessing the storage map from the memory and using the storage map to reconstruct the plurality of digital image files from the blocks of the decompressed new digital image files.

7. The computer-implemented method of claim 1 , wherein the pixel data comprises one or more of color data, luminance data, chrominance data, and hue data.

8. A computer-implemented method to perform preprocessing, compression, and archiving of a collection of digital image files comprising pixels, the method comprising:
with a processor, receiving a plurality of digital image files from a collection of digital image files;
extracting, with the processor, blocks from each of the digital image files wherein each block comprises a contiguous subset of the pixels of the respective image file;
analyzing, with the processor, each of the blocks to determine a corresponding frequency of pixel data variation for each of the blocks;
creating, with the processor, a plurality of new digital image files each comprising a plurality of the blocks extracted from at least two of the received digital image files, wherein the blocks selected for each particular new digital image file are those of the blocks of a given range of frequencies of pixel data variation;
generating, with the processor, a storage map that identifies a location of each block in the plurality of digital image files from a collection of digital image files and that identifies a corresponding location of each block in the new digital image files;
compressing, with the processor, each respective new digital image file using a predetermined image compression algorithm; and
with a processor, creating an image archive comprising the plurality of compressed new digital image files and the generated storage map, and storing the image archive to a memory.

9. The computer-implemented method of claim 8 , further comprising:
identifying, with the processor, a level of image compression to apply to each particular new digital image file based on the frequencies of pixel data variation of the selected blocks comprised by each respective new digital image file, wherein the identified level of image compression to apply to each particular new digital image file indicates a compression setting of the predetermined image compression algorithm, the compression setting corresponding to a compression ratio; and
applying, with the processor, the predetermined image compression algorithm to each particular new digital image file using the corresponding identified level of image compression.

10. The computer-implemented method of claim 9 , wherein identifying the level of image compression comprises:
for the new digital image files comprised of blocks having low frequencies of pixel data variation in comparison to other new digital image files, identifying, with the processor, a low compression setting to apply to each corresponding new digital image file to generate a compressed digital image file having relatively high image quality in comparison to other digital image files generated from new digital image files comprised of blocks having relatively high frequencies of pixel data variation; and
for the new digital image files comprised of blocks having high frequencies of pixel data variation in comparison to other new digital image files, identifying, with the processor, a high compression setting to apply to each corresponding new digital image file to generate a compressed digital image file having relatively low image quality in comparison to other digital image files generated from new digital image files comprised of blocks having relatively low fr

Assignees

• Leroy Luc
• Adobe Systems Inc

Inventors

• Leroy Luc

Classifications

• H04N19/60

  using transform coding · CPC title

• H04N19/14

  Coding unit complexity, e.g. amount of activity or edge presence estimation (H04N19/146 takes precedence) · CPC title

• H04N19/176

  the region being a block, e.g. a macroblock · CPC title

• H04N19/115 Primary

  Selection of the code volume for a coding unit prior to coding · CPC title

• H04N19/172

  the region being a picture, frame or field · CPC title