Generators

Generators produce data and have at least one output but no input.

File reader

class read

The reader loads single files from disk to produce a stream of two-dimensional data items. Supported file types depend on the compiled plugin. Raw (.raw) and EDF (.edf) files can always be read without additional support. Additionally, loading TIFF (.tif and .tiff) and HDF5 (.h5) files might be supported.

The nominal resolution can be decreased by specifying the y coordinate and a height. Due to reduced I/O, this can dramatically improve performance.

"path": string: Glob-style pattern that describes the file path. For HDF5 files this must point to a file and a data set separated by a colon, e.g. /path/to/file.h5:/my/data/set.

"number": uint: Number of files to read.

"start": uint: First index from where files are read.

"image-start": uint: First image index from where images are read in multi-image files.

"step": uint: Number of files to skip.

"image-step": uint: Number of images to skip in a multi-image file.

"y": uint: Vertical coordinate from where to start reading.

"height": uint: Height of the region that is read from the image.

"y-step": uint: Read every y-step row.

"convert": boolean: Convert input data to float elements, enabled by default.

"raw-width": uint: Specifies the width of raw files.

"raw-height": uint: Specifies the height of raw files.

"raw-bitdepth": uint: Specifies the bit depth of raw files.

"raw-pre-offset": ulong: Offset that is skipped before reading the next frame from the current file.

"raw-post-offset": ulong: Offset that is skipped after reading the last frame from the current file.

"type": enum: Overrides the type detection that is based on the file extension. For example, to load .foo files as raw files, set the type property to raw.

Memory reader

class memory-in

Reads data from a pre-allocated memory region. Unlike input and output tasks this can be used to interface with other code more directly, e.g. to read from a NumPy buffer or an OpenCL buffer:

import gi
import numpy as np
import tifffile
gi.require_version("Ufo", "0.0")
from gi.repository import Ufo


USE_OPENCL = True


pm = Ufo.PluginManager()
sched = Ufo.FixedScheduler()
graph = Ufo.TaskGraph()

mem_in = pm.get_task("memory-in")
writer = pm.get_task("write")

n = 4096
mem_in.props.width = n
mem_in.props.height = n
mem_in.props.bitdepth = 32
mem_in.props.number = 1

writer.props.filename = "out.tif"
writer.props.tiff_bigtiff = False

if USE_OPENCL:
    import pyopencl as cl
    import pyopencl.array as pa

    res = Ufo.Resources()
    # Use UFO's OpenCL context
    ctx = cl.Context.from_int_ptr(res.get_context())
    queue = cl.CommandQueue(ctx, ctx.devices[0])
    # And make sure UFO does not create a new one
    sched.set_resources(res)

    ref_buf = pa.arange(queue, n ** 2, dtype=np.float32).reshape(n, n)
    mem_in.props.memory_location = "buffer"
    mem_in.props.pointer = ref_buf.data.int_ptr
    ref = ref_buf.get()  # For checking later
else:
    ref = np.arange(n ** 2, dtype=np.float32).reshape(n, n)
    mem_in.props.memory_location = "host"
    mem_in.props.pointer = ref.__array_interface__["data"][0]

graph.connect_nodes(mem_in, writer)
sched.run(graph)

out = tifffile.imread("out.tif")
assert np.sum(out - ref) == 0.0

"pointer": ulong: Pointer to pre-allocated memory.

"width": uint: Specifies the width of input.

"height": uint: Specifies the height of input.

"number": uint: Specifies the number of items to read.

"complex-layout": boolean: Treat input as interleaved complex64 data type (x[0] = Re(z[0]), x[1] = Im(z[0]), …)

"memory-location": enum: Location of the input memory [host (RAM), buffer (OpenCL buffer)]

ZeroMQ subscriber

class zmq-sub

Generates a stream from a compatible ZeroMQ data stream, for example published by the zmq-pub task.

"address": string: Host address of the ZeroMQ publisher. Note, that as of now the publisher binds to a tcp endpoint, thus you have to use that as well. By default, the address is set to the local host address 127.0.0.1.

UcaCamera reader

class camera

The camera task uses libuca to read frames from a connected camera and provides them as a stream.

When name is provided, the corresponding plugin is instantiated by the camera task itself. However, an already configured UcaCamera object can also be passed via camera.

"name": string: Name of the camera that is used.

"number": uint: Number of frames that are recorded.

"properties": string: Property string, i.e. roi-width=512 exposure-time=0.1.

Note

This requires third-party library libuca.

stdin reader

class stdin

Reads data from stdin to produce a valid data stream. width, height and bitdepth must be set correctly to ensure correctly sized data items.

"width": uint: Specifies the width of input.

"height": uint: Specifies the height of input.

"bitdepth": uint: Specifies the bit depth of input.

"convert": boolean: Convert input data types to float, enabled by default.

Metaball simulation

class metaballs

Generate animated meta balls. In each time step the meta balls move by a random velocity.

"width": uint: Width of output data stream.

"height": uint: Height of output data stream.

"number-balls": uint: Number of meta balls.

"number": uint: Length of data stream.

Data generation

class dummy-data

Only asks for image data sized width times height times depth and forwards number of them to the next filter. The data is never touched if init is not set, thus it might be suitable for performance measurements.

"width": uint: Width of image data stream.

"height": uint: Height of image data stream.

"depth": uint: Depth of image data stream.

"number": uint: Number of images to produce.

"init": float: Value to initialize the output buffer.