Module: external.tifffile¶

Bases: object

Read image and metadata from TIFF, STK, LSM, and FluoView files.

TiffFile instances must be closed using the ‘close’ method, which is automatically called when using the ‘with’ context manager.

Examples

>>> with TiffFile('temp.tif') as tif:
...     data = tif.asarray()
...     data.shape
(5, 301, 219)

Attributes

__init__(arg, name=None, offset=None, size=None, multifile=True, multifile_close=True, pages=None, fastij=True, is_ome=None)[source]¶

Initialize instance from file.

Parameters:

arg : str or open file Name of file or open file object. The file objects are closed in TiffFile.close(). name : str Optional name of file in case ‘arg’ is a file handle. offset : int Optional start position of embedded file. By default this is the current file position. size : int Optional size of embedded file. By default this is the number of bytes from the ‘offset’ to the end of the file. multifile : bool If True (default), series may include pages from multiple files. Currently applies to OME-TIFF only. multifile_close : bool If True (default), keep the handles of other files in multifile series closed. This is inefficient when few files refer to many pages. If False, the C runtime may run out of resources. pages : sequence of int Indices of the pages to read. If None (default) all pages are read. Can be used to read only the first page with pages=[0]. Specifying pages might invalidate series based on metadata. fastij : bool If True (default), try to use only the metadata from the first page of ImageJ files. Significantly speeds up loading movies with thousands of pages. is_ome : bool If False, disable processing of OME-XML metadata.

asarray(key=None, series=None, memmap=False, tempdir=None)[source]¶

Return image data from multiple TIFF pages as numpy array.

By default the first image series is returned.

Parameters:

key : int, slice, or sequence of page indices Defines which pages to return as array. series : int or TiffPageSeries Defines which series of pages to return as array. memmap : bool If True, return an read-only array stored in a binary file on disk if possible. The TIFF file is used if possible, else a temporary file is created. tempdir : str The directory where the memory-mapped file will be created.

close()[source]¶

Close open file handle(s).

filehandle¶

Return file handle.

filename¶

Return name of file handle.

fstat¶

Lazy object attribute whose value is computed on first access.

info(series=None, pages=None)[source]¶

Return string with detailed information about file.

is_bigtiff¶

Lazy object attribute whose value is computed on first access.

is_fei¶

Lazy object attribute whose value is computed on first access.

is_fluoview¶

Lazy object attribute whose value is computed on first access.

is_imagej¶

Lazy object attribute whose value is computed on first access.

is_indexed¶

Lazy object attribute whose value is computed on first access.

is_lsm¶

Lazy object attribute whose value is computed on first access.

is_mdgel¶

Lazy object attribute whose value is computed on first access.

is_mediacy¶

Lazy object attribute whose value is computed on first access.

is_micromanager¶

Lazy object attribute whose value is computed on first access.

is_nih¶

Lazy object attribute whose value is computed on first access.

is_ome¶

Lazy object attribute whose value is computed on first access.

is_rgb¶

Lazy object attribute whose value is computed on first access.

is_scn¶

Lazy object attribute whose value is computed on first access.

is_sem¶

Lazy object attribute whose value is computed on first access.

is_stk¶

Lazy object attribute whose value is computed on first access.

is_tvips¶

Lazy object attribute whose value is computed on first access.

is_vista¶

Lazy object attribute whose value is computed on first access.

series¶

Lazy object attribute whose value is computed on first access.

Bases: object

Sequence of image files.

The data shape and dtype of all files must match.

Attributes

__init__(files, imread=<class 'skimage.external.tifffile.tifffile.TiffFile'>, pattern='axes', *args, **kwargs)[source]¶

Initialize instance from multiple files.

Parameters:	files : str, or sequence of str Glob pattern or sequence of file names. Binary streams are not supported. imread : function or class Image read function or class with asarray function returning numpy array from single file. pattern : str Regular expression pattern that matches axes names and sequence indices in file names. By default this matches Olympus OIF and Leica TIFF series.

exception ParseError[source]¶

Bases: Exception

asarray(memmap=False, tempdir=None, *args, **kwargs)[source]¶

Read image data from all files and return as single numpy array.

If memmap is True, return an array stored in a binary file on disk. The args and kwargs parameters are passed to the imread function.

Raise IndexError or ValueError if image shapes do not match.

close()[source]¶

Bases: object

Write image data to TIFF file.

TiffWriter instances must be closed using the ‘close’ method, which is automatically called when using the ‘with’ context manager.

Examples

>>> data = numpy.random.rand(2, 5, 3, 301, 219)
>>> with TiffWriter('temp.tif', bigtiff=True) as tif:
...     for i in range(data.shape[0]):
...         tif.save(data[i], compress=6)

__init__(file, append=False, bigtiff=False, byteorder=None, software='tifffile.py', imagej=False)[source]¶

Open a TIFF file for writing.

Existing files are overwritten by default. Use bigtiff=True when creating files larger than 2 GB.

Parameters:

file : str, binary stream, or FileHandle File name or writable binary stream, such as a open file or BytesIO. The file is created if it does not exist. append : bool If True and ‘file’ is an existing standard TIFF file, image data and tags are appended to the file. Appending data may corrupt specifically formatted TIFF files such as LSM, STK, ImageJ, NIH, or FluoView. bigtiff : bool If True, the BigTIFF format is used. byteorder : {‘<’, ‘>’} The endianness of the data in the file. By default this is the system’s native byte order. software : str Name of the software used to create the file. Saved with the first page in the file only. imagej : bool If True, write an ImageJ hyperstack compatible file. This format can handle data types uint8, uint16, or float32 and data shapes up to 6 dimensions in TZCYXS order. RGB images (S=3 or S=4) must be uint8. ImageJ’s default byte order is big endian but this implementation uses the system’s native byte order by default. ImageJ does not support BigTIFF format or LZMA compression. The ImageJ file format is undocumented.

TAGS = {'photometric': 262, 'planar_configuration': 284, 'smin_sample_value': 340, 'strip_offsets': 273, 'extra_samples': 338, 'color_map': 320, 'y_resolution': 283, 'tile_byte_counts': 325, 'predictor': 317, 'tile_depth': 32998, 'samples_per_pixel': 277, 'page_name': 285, 'image_width': 256, 'compression': 259, 'strip_byte_counts': 279, 'subfile_type': 255, 'software': 305, 'sample_format': 339, 'new_subfile_type': 254, 'tile_offsets': 324, 'bits_per_sample': 258, 'resolution_unit': 296, 'image_depth': 32997, 'image_description': 270, 'x_resolution': 282, 'datetime': 306, 'smax_sample_value': 341, 'image_length': 257, 'tile_length': 323, 'tile_width': 322, 'orientation': 274, 'document_name': 269, 'rows_per_strip': 278}¶

TYPES = {'H': 3, '2I': 5, '2i': 10, 'h': 8, 'd': 12, 'b': 6, 'I': 4, 'B': 1, 'i': 9, 'q': 17, 's': 2, 'Q': 16, 'f': 11}¶

close(truncate=False)[source]¶

Write remaining pages (if not truncate) and close file handle.

save(data, photometric=None, planarconfig=None, tile=None, contiguous=True, compress=0, colormap=None, description=None, datetime=None, resolution=None, metadata={}, extratags=())[source]¶

Write image data and tags to TIFF file.

Image data are written in one stripe per plane by default. Dimensions larger than 2 to 4 (depending on photometric mode, planar configuration, and SGI mode) are flattened and saved as separate pages. The ‘sample_format’ and ‘bits_per_sample’ tags are derived from the data type.

Parameters:

data : numpy.ndarray Input image. The last dimensions are assumed to be image depth, height (length), width, and samples. If a colormap is provided, the dtype must be uint8 or uint16 and the data values are indices into the last dimension of the colormap. photometric : {‘minisblack’, ‘miniswhite’, ‘rgb’, ‘palette’, ‘cfa’} The color space of the image data. By default this setting is inferred from the data shape and the value of colormap. For CFA images, DNG tags must be specified in extratags. planarconfig : {‘contig’, ‘planar’} Specifies if samples are stored contiguous or in separate planes. By default this setting is inferred from the data shape. If this parameter is set, extra samples are used to store grayscale images. ‘contig’: last dimension contains samples. ‘planar’: third last dimension contains samples. tile : tuple of int The shape (depth, length, width) of image tiles to write. If None (default), image data are written in one stripe per plane. The tile length and width must be a multiple of 16. If the tile depth is provided, the SGI image_depth and tile_depth tags are used to save volume data. Few software can read the SGI format, e.g. MeVisLab. contiguous : bool If True (default) and the data and parameters are compatible with previous ones, if any, the data are stored contiguously after the previous one. Parameters ‘photometric’ and ‘planarconfig’ are ignored. compress : int or ‘lzma’ Values from 0 to 9 controlling the level of zlib compression. If 0, data are written uncompressed (default). Compression cannot be used to write contiguous files. If ‘lzma’, LZMA compression is used, which is not available on all platforms. colormap : numpy.ndarray RGB color values for the corresponding data value. Must be of shape (3, 2**(data.itemsize*8)) and dtype uint16. description : str The subject of the image. Saved with the first page only. Cannot be used with the ImageJ format. datetime : datetime Date and time of image creation. Saved with the first page only. If None (default), the current date and time is used. resolution : (float, float[, str]) or ((int, int), (int, int)[, str]) X and Y resolutions in pixels per resolution unit as float or rational numbers. A third, optional parameter specifies the resolution unit, which must be None (default for ImageJ), ‘inch’ (default), or ‘cm’. metadata : dict Additional meta data to be saved along with shape information in JSON or ImageJ formats in an image_description tag. If None, do not write a second image_description tag. extratags : sequence of tuples Additional tags as [(code, dtype, count, value, writeonce)]. code : int The TIFF tag Id. dtype : str Data type of items in ‘value’ in Python struct format. One of B, s, H, I, 2I, b, h, i, 2i, f, d, Q, or q. count : int Number of data values. Not used for string values. value : sequence ‘Count’ values compatible with ‘dtype’. writeonce : bool If True, the tag is written to the first page only.