Preventing feature loss when compressing images

1. A computer-implemented method comprising:

compressing, by one or more computer processors, an image while retaining pattern and feature information at original coordinates, wherein the image is a full resolution 3D image, comprising:

decomposing, by one or more computer processors, the image into one or more space-filling shapes, wherein each shape of the one or more space-filling shapes comprises a plurality of voxels;

unrolling, by one or more computer processors, each voxel in the plurality of voxels contained in each segmented space-filling shape into a flatten list containing pixels, wherein pixels appearing sooner in the flatten list are closer to a center of a respective voxel;

creating, by one or more computer processors, one or more image subsets based on one or more determined regions of interest identified by utilizing a first pass selection of the regions of interest based on the compressed image; and

training, by one or more computer processors, one or more models based the created one or more image subsets that contain one or more regions of interest based on the compressed image.

2. The method of claim 1 , further comprising:

generating, by one or more computer processors, a group of binary values comprising at least a binary value for a respective pixel by applying a contrast threshold to the respective pixel for each unrolled voxel in the plurality of voxels;

identifying, by one or more computer processors, one or binary patterns for the generated group of binary values;

reversibly calculating, by one or more computer processors, an index for each identified binary pattern that identifies a unique pattern of voxels and component pixels; and

generating, by one or more computer processors, the compressed image only containing calculated indexes for each identified pattern.

3. The method of claim 2 , wherein generating the compressed image only containing calculated indexes for each identified pattern, comprises:

replacing, by one or more computer processors, an identified pattern with a corresponding calculated index.

4. The method of claim 2 , wherein identifying one or binary patterns for the generated group of binary values utilizes cross correlation to identify one or more binary patterns that have a high pattern correlation.

5. The method of claim 1 , wherein the one or more models are convolutional neural networks.

6. The method of claim 5 , further comprising:

deploying, by one or more computer processors, the one or more convolutional neural networks.

7. The method of claim 1 , wherein the space-filling shapes are polyhedrons.

8. A computer program product comprising:

one or more computer readable storage media and program instructions stored on the one or more computer readable storage media, the stored program instructions comprising:

program instructions to compress an image while retaining pattern and feature information at original coordinates, wherein the image is a full resolution 3D image, comprising:

program instructions to decompose the image into one or more space-filling shapes, wherein each shape of the one or more space-filling shapes comprises a plurality of voxels;

program instructions to unroll each voxel in the plurality of voxels contained in each segmented space-filling shape into a flatten list containing pixels, wherein pixels appearing sooner in the flatten list are closer to a center of a respective voxel;

program instructions to create one or more image subsets based on the one or more determined regions of interest identified by utilizing a first pass selection of the regions of interest based on the compressed image; and

program instructions to train one or more models based the created one or more image subsets that contain one or more regions of interest based on the compressed image.

9. The computer program product of claim 8 , wherein the program instructions, stored on the one or more computer readable storage media, further comprise:

program instructions to generate a group of binary values comprising at least a binary value for a respective pixel by applying a contrast threshold to the respective pixel for each unrolled voxel in the plurality of voxels;

program instructions to identify one or binary patterns for the generated group of binary values;

program instructions to reversibly calculate an index for each identified binary pattern that identifies a unique pattern of voxels and component pixels; and

program instructions to generate the compressed image only containing calculated indexes for each identified pattern.

10. The computer program product of claim 9 , wherein the program instructions to generate the compressed image only containing calculated indexes for each identified pattern, comprise:

program instructions to replace an identified pattern with a corresponding calculated index.

11. The computer program product of claim 9 , wherein identifying one or binary patterns for the generated group of binary values utilizes cross correlation to identify one or more binary patterns that have a high pattern correlation.

12. The computer program product of claim 8 , wherein the one or more models are convolutional neural networks.

13. A computer system comprising:

one or more computer processors;

one or more computer readable storage media; and

program instructions stored on the computer readable storage media for execution by at least one of the one or more processors, the stored program instructions comprising:

program instructions to compress an image while retaining pattern and feature information at original coordinates, wherein the image is a full resolution 3D image, comprising:

program instructions to decompose the image into one or more space-filling shapes, wherein each shape of the one or more space-filling shapes comprises a plurality of voxels;

program instructions to unroll each voxel in the plurality of voxels contained in each segmented space-filling shape into a flatten list containing pixels, wherein pixels appearing sooner in the flatten list are closer to a center of a respective voxel;

program instructions to create one or more image subsets based on the one or more determined regions of interest identified by utilizing a first pass selection of the regions of interest based on the compressed image; and

program instructions to train one or more models based the created one or more image subsets that contain one or more regions of interest based on the compressed image.

14. The computer program product of claim 13 , wherein the program instructions, stored on the one or more computer readable storage media, further comprise:

program instructions to generate a group of binary values comprising at least a binary value for a respective pixel by applying a contrast threshold to the respective pixel for each unrolled voxel in the plurality of voxels;

program instructions to identify one or binary patterns for the generated group of binary values;

program instructions to reversibly calculate an index for each identified binary pattern that identifies a unique pattern of voxels and component pixels; and

program instructions to generate the compressed image only containing calculated indexes for each identified pattern.

15. The computer system of claim 14 , wherein the program instructions to generate the compressed image only containing calculated indexes for each identified pattern, comprise:

program instructions to replace an identified pattern with a corresponding calculated index.

16. The computer system of claim 14 , wherein identifying one or binary patterns for the generated group of binary values utilizes cross correlation to identify one or more binary patterns that have a high pattern correlation.

17. The computer system of claim 13 , wherein the one or more models are convolutional neural networks.