OmeTiffSaver

class io.savers.OmeTiffSaver

Bases: io.savers.BaseSaver

OMETIFFSAVER - Saver for OME-TIFF (Open Microscopy Environment TIFF) output.

Handles two format variants: ‘OME-TIFF 5D (.ome.tiff)’ - full 5-D OME-TIFF (single file) ‘OME-TIFF 2D sequence (.ome.tiff)’ - one OME-TIFF per Z×T slice

OME-TIFF stores the full 5-D data set [H, W, C, D, T] together with standardised OME-XML metadata describing pixel sizes, channel names, LUT colours, and acquisition information.

The saver calls io.BioFormats.mibImage2ometiff(), which is already present in MIB3 at mib/+io/+BioFormats/mibImage2ometiff.m.

DATA DIMENSIONS Input data : [H, W, D, C, T] (MIB3 native order) mibImage2ometiff() expects [H, W, C, D, T] - the saver permutes dimensions 3 and 4 before the call.

SAVING OPTIONS PASSED TO mibImage2ometiff savingOptions.pixSize - struct {.x .y .z .units .t .tunits} savingOptions.lutColors - [C x 3] channel LUT colours (0..1) savingOptions.ImageDescription - (char) dataset description savingOptions.DimensionOrder - always ‘XYCZT’ savingOptions.Saving3d - ‘5D’ | ‘2D’ savingOptions.overwrite - logical savingOptions.DatasetType - ‘image’ | ‘mask’ | ‘labels’ savingOptions.showWaitbar - logical

NOTES * OME-TIFF is the preferred format for multichannel, multi-Z, multi-time datasets because it stores all metadata in standardised OME-XML. * The output file extension is always ‘.ome.tiff’; if the user provides a different extension it is replaced automatically.

USAGE EXAMPLES

%% 1. Save 5-D multichannel stack as OME-TIFF
saver = io.SaverFactory.create('OME-TIFF 5D (*.ome.tiff)');

opts.Format         = 'OME-TIFF 5D (*.ome.tiff)';
opts.showWaitbar    = false;
opts.silent         = true;
opts.overwrite      = true;
opts.layerType      = 'image';

meta.filename       = 'source_stack.tif';
meta.colorType      = 'multichannel';
meta.lutColors      = [1 0 0; 0 1 0; 0 0 1];   % R, G, B channels
meta.dataClass      = 'uint16';
meta.maxInt         = 65535;
meta.pixSize        = struct('x',0.065,'y',0.065,'z',0.2, ...
                             'units','um','t',1,'tunits','s');
meta.imageDescription = 'My confocal dataset';

data = uint16(rand(512,512,50,3,4)*65535);  % [H W D C T]
fnOut = saver.save(data, meta, '/output/myStack.ome.tiff', opts);
fprintf('Saved: %s\n', fnOut);

%% 2. Save as OME-TIFF 2D sequence
saver = io.SaverFactory.create('OME-TIFF 2D sequence (*.ome.tiff)');

opts.Format      = 'OME-TIFF 2D sequence (*.ome.tiff)';
opts.showWaitbar = true;
opts.silent      = true;
opts.overwrite   = true;
opts.layerType   = 'image';

meta.filename    = 'source_stack.tif';
meta.colorType   = 'multichannel';
meta.lutColors   = [1 0 0; 0 1 0; 0 0 1];
meta.dataClass   = 'uint16';
meta.maxInt      = 65535;
meta.pixSize     = struct('x',0.065,'y',0.065,'z',0.2, ...
                          'units','um','t',1,'tunits','s');

data = uint16(rand(512,512,50,3,1)*65535);  % [H W D C T]
fnOut = saver.save(data, meta, '/output/myStack.ome.tiff', opts);

%% 3. Via MibModel batch
BatchOpt.LayerType       = {'image'};
BatchOpt.Format          = {'OME-TIFF 5D (*.ome.tiff)'};
BatchOpt.OutputDirectoryPolicy = {'Full path'};
BatchOpt.DestinationDirectory  = '/output/dir';
BatchOpt.FilenamePolicy  = {'Use existing name'};
BatchOpt.showWaitbar     = false;
BatchOpt.mibBatchTooltip.LayerType = '';
model.save('image', [], BatchOpt);

SEE ALSO io.SaverFactory, io.savers.BaseSaver, io.savers.TiffSaver, io.BioFormats.mibImage2ometiff, core.MibImage.save, core.MibDataset.save, models.MibModel.save

Constructor Summary
OmeTiffSaver(options)

OMETIFFSAVER - Constructor for OmeTiffSaver class.

Syntax:
saver = io.savers.OmeTiffSaver(options)
Input Arguments:

  • options - (optional) struct, saver-level options (usually empty; per-save options are passed to save() instead)

Output Arguments:

  • obj - instance of the OmeTiffSaver class

Method Summary
getSupportedFormats(~)

GETSUPPORTEDFORMATS - Return format strings handled by OmeTiffSaver.

Syntax:
formats = obj.getSupportedFormats()
Input Arguments:

(none)

Output Arguments:

  • formats - cell array of format strings for OME-TIFF output

save(data, metadata, filename, options)

SAVE - Write data as an OME-TIFF file or 2-D OME-TIFF sequence.

Syntax:
fnOut = obj.save(data, metadata, filename, options)
Input Arguments:

  • data - [H, W, D, C, T] numeric array

  • metadata - struct with fields:

    • colorType - 'grayscale' | 'multichannel' | 'indexed'

    • lutColors - [C × 3] per-channel LUT colours (0-1 range)

    • dataClass - 'uint8' | 'uint16' | …

    • maxInt - maximum intensity value

    • pixSize - struct {.x, .y, .z, .units, .t, .tunits}

    • imageDescription - (optional) [char] dataset description string

    • sliceName - (optional) per-slice source filenames (used in 2D mode)

  • filename - [char] full output path; extension is always normalized to .ome.tiff

  • options - struct with fields:

    • Format - format string ('OME-TIFF 5D (*.ome.tiff)' or 'OME-TIFF 2D sequence (*.ome.tiff)')

    • layerType - 'image' | 'mask' | 'labels'; default: 'image'

    • showWaitbar - logical; default: true

    • silent - logical, suppress dialogs; default: false

    • overwrite - logical; default: true

    • FilenameGenerator - 'Use original filename' | 'Use sequential filename' (2D mode only)

Output Arguments:

  • fnOut - [char] path of saved .ome.tiff file, [] on failure

Example - see class-level documentation above.

saveStream(provider, metadata, filename, options)

SAVESTREAM - Stream an OME-TIFF from a SliceProvider (memory-bounded).

Syntax:
fnOut = obj.saveStream(provider, metadata, filename, options)

Memory-bounded twin of save: instead of receiving a full [H W D C T] array it pulls each Z-slice from provider.getSlice(z, t) (an io.savers.SliceProvider) so a large pyramid level is never gathered whole - peak memory stays at one XY slice.

  • 5D mode - writes a single OME-TIFF, streaming one plane at a time through the Bio-Formats Java writer (loci.formats.ImageWriter / OMETiffWriter.saveBytes). OME-XML metadata is built from the provider’s dimensions (MetadataTools.populateMetadata), so no full array is needed.

  • 2D sequence mode - writes one .ome.tiff per Z×T slice via imwrite.

NOTE: streaming bounds memory along Z. A single very large XY plane (e.g. a gigapixel WSI at full resolution, Z=1) is still held whole; tiled BigTIFF output for that case is separate future work.

Input/Output: see io.savers.BaseSaver.saveStream.