Image¶

This module wraps various parts of the Pillow image library to better integrate it with other components of arlunio in order to help produce and manipulate images that can be saved eventually to disk.

Creating Images¶

arlunio.image.new(size, *args, mode='RGBA', **kwargs) → arlunio.image.Image [source]¶

Creates a new image with the given size.

This function by default will return a new RGBA image with the given dimensions. Dimensions can be specified either using a tuple (width, height) or by passing in width and height individually as positional parameters.

This makes use of pillow’s PIL.Image.new() function, additional keyword arguments passed to this function will be passed onto it.

Parameters

size – The dimensions of the image, (width, height)
mode – The type of image to create, default RGBA. See Modes for more details.

arlunio.image.fromarray(*args, **kwargs)[source]¶

Create an image from an array

See PIL.Image.fromarray()

Image I/O¶

arlunio.image.encode(image: arlunio.image.Image) → bytes [source]¶

Return the image encoded as a base64 string.

Parameters: image – The image to encode.

Example

>>> import arlunio.image as image
>>> img = image.new((8, 8), color='red')
>>> image.encode(img)
b'iVBORw0KGgoAAAANSUhEUgAAAAgAAAAICAYAAADED76LAAAAFklEQVR4nGP8z8DwnwEPYMInOXwUAAASWwIOH0pJXQAAAABJRU5ErkJggg=='

arlunio.image.decode(bytestring: bytes) → arlunio.image.Image [source]¶

Decode the image represented by the given bytestring into an image object.

Parameters: bytestring – The bytestring to decode.

Example

import arlunio.image as image

bytestring = b'iVBORw0KGgoAAAANSUhEUgAAAAgAAAAICAYAAADED76LAAAAFklEQVR4nGP8z8DwnwEPYMInOXwUAAASWwIOH0pJXQAAAABJRU5ErkJggg=='  # noqa: E501
img = image.decode(bytestring)

arlunio.image.load(*args, **kwargs) → arlunio.image.Image [source]¶

Load an image from the given file.

See PIL.Image.open()

arlunio.image.save(image: arlunio.image.Image, filename: str, mkdirs: bool = False) → None [source]¶

Save an image in PNG format.

Parameters

filename – The filepath to save the image to.
mkdirs – If true, make any parent directories

Manipulating Images¶

arlunio.image.colorramp(values, start: Optional[str] = None, stop: Optional[str] = None) → arlunio.image.Image [source]¶

Given a 2d array of values, produce an image gradient based on them.

First this function will scale the input array so that all values fall in the range \([0, 1]\). It will then produce an image with the same dimensions as the original array. The color of each pixel will be chosen based on the corresponding value of the scaled array.

If the value is \(0\) the color will be given by the start parameter
If the value is \(1\) the color will be given by the stop parameter
Otherwise the color will be some mix between the two.

Parameters

values – The array of values used to decide on the color.
start – The color to use for values near \(0\) (default, black)
stop – The color to use for values near \(1\) (default, white)

Examples

import arlunio.image as image
import arlunio.math as math
import numpy as np

cartesian = math.Cartesian()
p = cartesian(width=256, height=256)

x = image.colorramp(p[:, :, 0], start="#0000", stop="#f007")
y = image.colorramp(p[:, :, 1], start="#0000", stop="#00f7")

img = x + y

arlunio.image.fill(mask: arlunio.mask.Mask, foreground: Optional[str] = None, background: Optional[str] = None, image: Optional[arlunio.image.Image] = None) → arlunio.image.Image [source]¶

Apply color to an image, as specified by a mask.

Parameters

mask – The mask that selects the region to be coloured
foreground – A string representation of the color to use, this can be in any format that is supported by the PIL.ImageColor module. If omitted this will default to black.
background – In the case where an existing image is not provided this parameter can be used to set the background color of the generated image. This can be any string that is accepted by the PIL.ImageColor module. If omitted this will default to transparent
image – The image to color in, if omitted a blank image will be used.

Example

import arlunio.image as image
import arlunio.shape as shape

circle = shape.Circle(x0=-0.5, y0=0.25, r=0.6)
img = image.fill(circle(width=512, height=256), foreground='red')

circle.x0, circle.y0 = 0, 0
img = image.fill(circle(width=512, height=256), foreground='#0f0', image=img)

circle.x0, circle.y0 = 0.5, -0.25
img = image.fill(circle(width=512, height=256), foreground='blue', image=img)