MibBigDataImage

class core.MibBigDataImage

Bases: core.@MibVirtualImage.MibVirtualImage

MIBBIGDATAIMAGE - BigData image class for MIB3 - on-demand reader for pyramidal, chunked datasets.

Subclass of core.MibVirtualImage. It inherits the full on-demand read machinery (getData → getDataZarr / getDataVirt, getOrCreateLoader, closeVirtualDataset) and adds the BigData identity used by core.MibDataset (datasetType = 'BigData').

Why a subclass of MibVirtualImage?

For browsing, BigData reads are identical to the Virtual zarr path - both stream sub-regions from a multi-resolution OME-Zarr v3 pyramid. What BigData adds on top (a disk-backed, writable, pyramidal segmentation model via io.adapters.ZarrBlockedAdapter, copy-or- modify image edits, and a Zarr3 saver) lives in core.MibDataset / core.MibBigDataLabels rather than in the image reader. Keeping the image reader as a thin subclass means isa(obj, 'core.MibVirtualImage') stays true, so reader-cleanup paths (e.g. closeVirtualDataset) work unchanged.

Type marker: obj.type = 'bigdata' (the parent sets 'virtual').

Constructor Summary
MibBigDataImage(data, meta)

MIBBIGDATAIMAGE - Construct a BigData image reader from a metadata dictionary.

Syntax:
obj = core.MibBigDataImage(data, meta)
obj = core.MibBigDataImage([], meta)

Delegates to core.MibVirtualImage, which opens the zarr store, populates obj.pyramid (levelNames, levelImageSizes, levelScaleFactors, chunkSizes, axisOrder) from the OME-NGFF multiscales metadata, and sets up the reader handle. This constructor then overrides obj.type to 'bigdata' so MibDataset routes label and export paths correctly.

Input Arguments:

  • data (optional) - [cell | empty] cell array of file path string(s) to the zarr3 dataset, or [] for a placeholder object with no open store.

  • meta (optional) - [dictionary] metadata dictionary produced by a setup loader (e.g. io.loaders.BioFormatsVirtualSetupLoader or io.loaders.Zarr3VirtualSetupLoader). Default: empty MibImage info.

Example - open a previously converted OME-Zarr pyramid as BigData:

opts = struct('datasetMode', 'BigData', 'silentMode', true, ...
              'mibPath', 'C:\MIB3\mib', 'ParentFigure', []);
loader = io.loaders.Zarr3VirtualSetupLoader(opts);
[meta, filelist] = loader.loadMetadata({'C:\data\slide.zarr3'}, opts);
[imgInfo, meta]  = loader.loadImages(filelist, meta, opts);
obj = core.MibBigDataImage(imgInfo, meta);
% obj.type == 'bigdata';  obj.pyramid.levelNames has N level paths