Dynamic digital image compression based on digital image characteristics

Patent metadata

Field	Value
Publication number	US-9800882-B2
Application number	US-201615352092-A
Country	US
Kind code	B2
Filing date	Nov 15, 2016
Priority date	Mar 16, 2015
Publication date	Oct 24, 2017
Grant date	Oct 24, 2017

Abstract

Systems and techniques are disclosed for dynamically and automatically selecting an appropriate compression technique and/or compression parameters for digital images in order to reduce or prevent loss of significant information that may negatively impact the utility or usefulness of the digital images. For example, based on various image characteristics associated with a digital image, the system may dynamically compress the image using particular compression techniques and/or by adjusting compression parameters, to maintain significant information of the image. The system may select compression techniques and/or compression parameters based on one or more compression rules, which may be associated with image characteristics, patient characteristics, medical history, etc. Further, the system may, based on the one or more compression rules, compress the image to a maximum degree of compression while maintaining the significant information of the image.

First claim

What is claimed is:

1. A method of digital image compression, the method comprising:
causing execution of software instructions by one or more hardware computing devices in order to:
receive an uncompressed digital image;
determine a characteristic associated with the uncompressed digital image;
access a compression rule associated with the characteristic;
compress, based on the compression rule, the uncompressed digital image using each of a plurality of compression levels to generate a set of compressed digital images that are each compressed at different compression levels;
for each of the compressed digital images of the set, determine a respective amount of errors based on comparison of the respective compressed digital images to the uncompressed digital image; and
determine a first compressed digital image of the set that:
is associated with an amount of errors that satisfies a threshold defined by the compression rule, and
is compressed at a highest compression level of the plurality of compression levels that has an amount of errors that satisfies the threshold defined by the compression rule.

2. The method of claim 1 , wherein the characteristic associated with the uncompressed digital image includes at least one of: an imaging modality, an anatomical feature, or an acquisition type.

3. The method of claim 1 , wherein the plurality of compression levels each includes a compression algorithm and a set of respective quality factors, and the set of compressed digital images is generated based on the respective compression algorithms and quality factors.

4. The method of claim 1 , wherein determining an amount of errors by comparison of a compressed digital image to the uncompressed digital image comprises:
causing execution of software instructions by one or more hardware computing devices in order to:
determine a difference between the uncompressed digital image and the compressed digital image to generate difference image data; and
determine the amount of errors by at least one of:
determining a number of pixels in the difference image data having a value indicative of an error, or
determining a degree of error in one or more pixels of the difference image data.

5. The method of claim 1 , wherein determining an amount of errors by comparison of a compressed digital image to the uncompressed digital image comprises:
causing execution of software instructions by one or more hardware computing devices in order to:
identify one or more pixels of the uncompressed digital image having an intensity value satisfying a threshold;
for each pixel of the one or more pixels, determine a difference between the pixel of the uncompressed digital image and the pixel of the compressed digital image to generate difference image data;
determine the amount of errors by at least one of:
determining a number of pixels in the difference image data having a value indicative of an error, or
determining a degree of error in one or more pixels of the difference image data.

6. The method of claim 5 , wherein each of the one or more pixels comprises a group of pixels, and wherein each group of pixels comprises at least one of: a 4×4 group of pixels, or a 6×6 group of pixels.

7. The method of claim 1 , wherein determining an amount of errors by comparison of a compressed digital image to the uncompressed digital image comprises:
causing execution of software instructions by one or more hardware computing devices in order to:
identify one or more regions of the uncompressed digital image having pixel intensity values satisfying a threshold;
for each pixel or group of pixels of the one or more regions, determine a difference between the pixel of the uncompressed digital image and the pixel of the compressed digital image to generate difference image data; and
determine the amount of errors by at least one of:
determining a number of pixels in the difference image data having a value indicative of an error, or
determining a degree of error in one or more pixels of the difference image data.

8. The method of claim 1 further comprising:
causing execution of software instructions by one or more hardware computing devices in order to:
store the first compressed digital image in a data store.

9. The method of claim 1 further comprising:
causing execution of software instructions by one or more hardware computing devices in order to:
in response to determining that none of the compressed digital images of the set is associated with an amount of errors that satisfies the threshold defined by the compression rule, recompress the uncompressed digital image using a lossless compression technique to generate a second compressed digital image; and
store the second compressed digital image in a data store.

10. The method of claim 1 , wherein the highest compression level is a compression level that requires a least amount of storage space as compared to other compression levels of the plurality of compression levels, while having the amount of errors that satisfies the threshold defined by the compression rule.

11. A system comprising:
a non-transitory computer readable storage medium having software instructions embodied thereon; and
one or more processors configured to execute the software instructions to cause the one or more processors to:
receive an uncompressed digital image;
determine a characteristic associated with the uncompressed digital image;
access a compression rule associated with the characteristic;
compress, based on the compression rule, the uncompressed digital image using each of a plurality of compression levels to generate a set of compressed digital images that are each compressed at different compression levels;
for each of the compressed digital images of the set, determine a respective amount of errors based on comparison of the respective compressed digital images to the uncompressed digital image; and
determine a first compressed digital image of the set that:
is associated with an amount of errors that satisfies a threshold defined by the compression rule, and
is compressed at a highest compression level of the plurality of compression levels that has an amount of errors that satisfies the threshold defined by the compression rule.

12. The system of claim 11 , wherein the characteristic associated with the uncompressed digital image includes at least one of: an imaging modality, an anatomical feature, or an acquisition type.

13. The system of claim 11 , wherein the plurality of compression levels each includes a compression algorithm and a set of respective quality factors, and the set of compressed digital images is generated based on the respective compression algorithms and quality factors.

14. The system of claim 11 , wherein determining an amount of errors by comparison of a compressed digital image to the uncompressed digital image comprises:
determining a difference between the uncompressed digital image and the compressed digital image to generate difference image data; and
determining the amount of errors by at least one of:
determining a number of pixels in the difference image data having a value indicative of an error, or
determining a degree of error in one or more pixels of the difference image data.

15. The system of claim 11 , wherein determining an amount of errors by comparison of a compressed digital image to the uncompressed digital image comprises:
identifying one or more pixels of the uncompressed digital image having an intensity value satisfying a threshold;
for each pixel of the one or more pixels, determining a diffe

Assignees

• D R Systems Inc

Inventors

• Fram Evan K

Classifications

• G06T2207/30004

  Biomedical image processing · CPC title

• G06T7/0002

  Inspection of images, e.g. flaw detection · CPC title

• G06F19/321

  Physics · mapped topic

• H04N19/124

  Quantisation · CPC title

• G06T2207/10081

  Computed x-ray tomography [CT] · CPC title