Proposals:HDF5 ImageIO: Difference between revisions
From KitwarePublic
Jump to navigationJump to search
Kishoreraom (talk | contribs) |
|||
| Line 15: | Line 15: | ||
* Multi-Channel images | * Multi-Channel images | ||
* Large datasets ( Size > 4Gb ) | * Large datasets ( Size > 4Gb ) | ||
* Megason Lab, Harvard University | |||
* Single experiment images of size 1024 x 1024 x 75 (XYZ), 2 channels, 1000 time-points | |||
* 8bit and 16bit | |||
* Images stored as 2D PNGs with filenames giving location | |||
* Need to support optimized reading (image streaming) of a sub-volume | |||
* Eg: Box filtering using a kernel of size 5x5x1x1x3 | |||
* Cyclic buffer optimization in the ITK reader that keeps overlapping data and only reads new data | |||
* Multi-resolution images for heirarchical registration of multiple experimental sets | |||
* Compression is not as important in the short term but will be needed in the long term | |||
Revision as of 17:30, 25 May 2010
Adding support for HDF5
Motivation
Provide better support for large datasets, particularly those used in microscopy.
Historical Note
- ITK provides support for MINC2, which in turn is based on HDF5.
Use Cases
- Chunking (streaming)
- Multi-Resolution
- Multi-Channel images
- Large datasets ( Size > 4Gb )
- Megason Lab, Harvard University
- Single experiment images of size 1024 x 1024 x 75 (XYZ), 2 channels, 1000 time-points
- 8bit and 16bit
- Images stored as 2D PNGs with filenames giving location
- Need to support optimized reading (image streaming) of a sub-volume
- Eg: Box filtering using a kernel of size 5x5x1x1x3
- Cyclic buffer optimization in the ITK reader that keeps overlapping data and only reads new data
- Multi-resolution images for heirarchical registration of multiple experimental sets
- Compression is not as important in the short term but will be needed in the long term
