# This file is part of Sympathy for Data.
# Copyright (c) 2017, Combine Control Systems AB
#
# Sympathy for Data is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, version 3 of the License.
#
# Sympathy for Data is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Sympathy for Data. If not, see <http://www.gnu.org/licenses/>.
from sympathy.api import node
from sympathy.api.nodeconfig import Ports, Tag, Tags
import numpy as np
from skimage import morphology
from sylib.imageprocessing.image import ImagePort
from sylib.imageprocessing.algorithm_selector import ImageFiltering_abstract
def alg_gaussian(par):
width = par['width'].value
height = par['height'].value
cx = par['x'].value
cy = par['y'].value
# vxx = par['varxx'].value
# vxy = par['varxy'].value
# vyy = par['varyy'].value
p = par['p'].value
sigma_x = par['sigma-x'].value
sigma_y = par['sigma-y'].value
rotation = par['rotation'].value
cos_t = np.cos(rotation)
sin_t = np.sin(rotation)
sin_2t = np.sin(2*rotation)
vxx = (cos_t * cos_t / (2 * sigma_x * sigma_x)
+ sin_t * sin_t / (2 * sigma_y * sigma_y))
vyy = (sin_t * sin_t / (2 * sigma_x * sigma_x)
+ cos_t * cos_t / (2 * sigma_y * sigma_y))
vxy = (- sin_2t / (4 * sigma_x * sigma_x)
+ sin_2t / (4 * sigma_y * sigma_y))
scale = par['scale'].value
xx = np.array([[x - cx for x in range(width)] for _ in range(height)])
yy = np.array([[y - cy for _ in range(width)] for y in range(height)])
return scale * np.exp(-np.power(vxx*xx*xx + 2.0*vxy*xx*yy + vyy*yy*yy, p))
[docs]class ImageGenerate(ImageFiltering_abstract, node.Node):
name = 'Generate Image'
author = 'Mathias Broxvall'
version = '0.1'
icon = 'image_generate.svg'
description = 'Generates an image or structuring element of a given size'
nodeid = 'syip.imagegenerate'
tags = Tags(Tag.ImageProcessing.IO)
algorithms = {
'empty': {
'description': 'Generates an empty image of a given size',
'width': 'Width of generated image',
'height': 'Height of generated image',
'channels': 'Number of channels in generated image',
'k': 'Value for all pixels in all channels',
'algorithm': lambda par: np.full((
par['height'].value,
par['width'].value,
par['channels'].value), par['k'].value)
},
'disk': {
'description': 'Generates an circular binary structuring element',
'size': 'Radius of the disk',
'algorithm': lambda par: morphology.disk(par['size'].value)
},
'diamond': {
'description': (
'Generates a diamond-shaped binary structuring element.\n\n'
'A pixel is part of the neighborhood if the city '
'block/Manhattan distance between it and the center of the'
'neighborhood is no greater than radius.'),
'size': 'Radius of the disk',
'algorithm': lambda par: morphology.diamond(par['size'].value)
},
'square': {
'description': (
'Generates a square-shaped binary structuring element.'),
'size': 'Size of the square',
'algorithm': lambda par: morphology.square(par['size'].value)
},
'star': {
'description': (
'Generates a star-shaped binary structuring element.\n'
'The star has 8 vertices and is an overlap of a square of '
'size 2n + 1 with its 45 degree rotated version. The slanted '
'sides are 45 or 135 degrees to the horizontal axis.'),
'size': 'Size "N" of the square',
'algorithm': lambda par: morphology.star(par['size'].value)
},
'octagon': {
'description': (
'Generates an octagon-shaped binary structuring element.'),
'size': 'Size of horizontal/vertial parts of the octagon',
'other size': 'Size of diagonal parts of the octagon',
'algorithm': (
lambda par: morphology.octagon(par['size'].value,
par['other size'].value))
},
'rectangle': {
'description': (
'Generates a rectangle-shaped binary structuring element.'),
'width': 'Width of rectangle',
'height': 'Height of rectangle',
'algorithm': lambda par: morphology.rectangle(par['height'].value,
par['width'].value)
},
'gaussian': {
'description': (
'Generates an elliptical Gaussian from the given variance '
'matrix.'),
'width': 'Width of rectangle',
'height': 'Height of rectangle',
'x': 'Center point along x axis',
'y': 'Center point along y axis',
'scale': 'Multiplier for output',
'sigma-x': 'Variation along first axis before rotation',
'sigma-y': 'Variation along first axis before rotation',
'rotation': 'Rotation given in radians',
'p': 'Super-gaussian exponent (default 1)',
'algorithm': alg_gaussian
},
}
options_list = [
'width', 'height', 'size', 'channels', 'k', 'other size',
'sigma-x', 'sigma-y', 'rotation', 'scale', 'x', 'y', 'p'
]
options_types = {
'width': int,
'height': int,
'sigma-x': float,
'sigma-y': float,
'rotation': float,
'scale': float,
'x': float,
'y': float,
'k': float,
'p': float,
'size': int,
'other size': int,
'channels': int,
}
options_default = {
'width': 16,
'height': 16,
'sigma-x': 0.5,
'sigma-y': 0.5,
'rotation': 0.0,
'scale': 1.0,
'x': 0.0,
'y': 0.0,
'k': 0.0,
'p': 1.0,
'size': 5,
'other size': 5,
'channels': 1,
}
parameters = node.parameters()
parameters.set_string(
'algorithm', value=next(iter(algorithms)), description='',
label='Algorithm')
ImageFiltering_abstract.generate_parameters(
parameters, options_types, options_default)
outputs = Ports([
ImagePort('Resulting image', name='output'),
])
__doc__ = ImageFiltering_abstract.generate_docstring(
description, algorithms, options_list, None, outputs)
def execute(self, node_context):
params = node_context.parameters
alg_name = params['algorithm'].value
alg = self.algorithms[alg_name]['algorithm']
result = alg(params)
node_context.output['output'].set_image(result)
