This repository has been archived on 2023-03-25. You can view files and clone it, but cannot push or open issues or pull requests.
mightyscape-1.1-deprecated/extensions/fablabchemnitz/image_triangulation/image_triangulation.py

248 lines
10 KiB
Python

#!/usr/bin/env python3
'''
Copyright (C) 2014 Nicola Romano', romano.nicola@gmail.com
version 0.1
0.1: first working version
------------------------------------------------------------------------
This program 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; either version 2 of the License, or
(at your option) any later version.
This program 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 this program; if not, write to the Free Software
Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
------------------------------------------------------------------------
'''
import base64
from io import BytesIO
import inkex
import os
from PIL import Image
from lxml import etree
import numpy as np
from scipy.spatial import Delaunay
from scipy.cluster.vq import kmeans2
import cv2
import urllib.request as urllib
class ImageTriangulation(inkex.EffectExtension):
def add_arguments(self, pars):
pars.add_argument("-n", "--num_points", type=int, default=100, help="Number of points to be sampled")
pars.add_argument("-m", "--edge_thresh_min", type=int, default=200, help="Minimum threshold for edge detection")
pars.add_argument("-M", "--edge_thresh_max", type=int, default=255, help="Maximum threshold for edge detection")
pars.add_argument("-c", "--add_corners", type=inkex.Boolean, default=0, help="Use corners for triangulation?")
pars.add_argument("-g", "--gradient_fill", type=inkex.Boolean, default=0, help="Fill triangles with gradient?")
pars.add_argument("-b", "--tab", default='', help="The tab of the interface")
def draw_SVG_path(self, points, closed, style, parent):
pathdesc = "M "
for p in points:
pathdesc = pathdesc + str(p[0]) + "," + str(p[1]) + " "
if closed == 1:
pathdesc = pathdesc + "Z"
path = etree.SubElement(parent, inkex.addNS('path','svg'), {'style' : str(inkex.Style(style)), 'd' : pathdesc})
return path
def checkImagePath(self, node):
"""Embed the data of the selected Image Tag element"""
xlink = node.get('xlink:href')
if xlink and xlink[:5] == 'data:':
# No need, data alread embedded
return
url = urllib.urlparse(xlink)
href = urllib.url2pathname(url.path)
# Primary location always the filename itself.
path = self.absolute_href(href or '')
# Backup directory where we can find the image
if not os.path.isfile(path):
path = node.get('sodipodi:absref', path)
if not os.path.isfile(path):
inkex.errormsg('File not found "{}". Unable to embed image.').format(path)
return
if (os.path.isfile(path)):
return path
def effect(self):
# Check we have something selected
if len(self.svg.selected) == 0:
inkex.errormsg("Please select an image.")
exit()
else:
# Check it is an image
for id, obj in self.svg.selected.items():
if obj.tag[len(obj.tag)-5:] != "image":
inkex.errormsg("The selected object (" + id + ") is not an image, skipping.")
continue
else:
self.path = self.checkImagePath(obj) # This also ensures the file exists
grpname = 'img_triangles'
# Make sure that the id/name is unique
index = 0
while (str(self.svg.get_ids()) in grpname):
grpname = 'axis' + str(index)
index = index + 1
grp_name = grpname
grp_attribs = {inkex.addNS('label','inkscape'):grp_name}
# The group to put everything in
grp = etree.SubElement(self.svg.get_current_layer(), 'g', grp_attribs)
# Find image size and position in Inkscape
try:
self.img_x_pos = float(obj.get("x"))
self.img_y_pos = float(obj.get("y"))
except:
self.img_x_pos = 0
self.img_y_pos = 0
self.img_width = float(obj.get("width"))
self.img_height = float(obj.get("height"))
if self.path is None: #check if image is embedded or linked
image_string = obj.get('{http://www.w3.org/1999/xlink}href')
# find comma position
i = 0
while i < 40:
if image_string[i] == ',':
break
i = i + 1
im = Image.open(BytesIO(base64.b64decode(image_string[i + 1:len(image_string)])))
else:
im = Image.open(self.path)
# IMPORTANT!
# The numpy array is accessed as im.data[row,column], that is data[y_coord, x_coord]
# Be careful not to pass coordinates as (x,y): rather use (y,x)!
im.data = np.asarray(im)
# The RGB components of all the pixels in the image
self.red, self.green, self.blue = im.data[:,:,0], im.data[:,:,1], im.data[:,:,2]
# Find real image size
(self.img_real_width, self.img_real_height) = im.size
self.doTriangulation(im, grp)
# Converts image coordinates to screen coordinates
def imgToScreen(self, x, y):
newx = x / (self.img_real_width/self.img_width) + self.img_x_pos
newy = y / (self.img_real_height/self.img_height) + self.img_y_pos
return (newx, newy)
def createLinearGradient(self, x1, y1, x2, y2, color1, color2, gradID):
attribs = {
'x1' : str(x1),
'y1' : str(y1),
'x2' : str(x2),
'y2' : str(y2),
'id' : gradID,
'gradientUnits' : "userSpaceOnUse",
'{'+inkex.NSS[u'xlink']+'}href': "#"+gradID
}
svgdefs = self.document.getroot().find(inkex.addNS('defs', 'svg'))
gradient = etree.SubElement(svgdefs, inkex.addNS('linearGradient','svg'), attribs)
attribs = {
'offset' : "0%",
'style' : "stop-color:"+color1+"; stop-opacity:1"
}
stop1 = etree.SubElement(gradient, inkex.addNS('stop','svg'), attribs)
attribs = {
'offset' : "100%",
'style' : "stop-color:"+color2+"; stop-opacity:1"
}
stop2 = etree.SubElement(gradient, inkex.addNS('stop','svg'), attribs)
return gradient
def doTriangulation (self, im, grp):
# Read image with OpenCV
imcv = np.array(im)
#imcv = cv2.imread(self.path)
# Convert to grayscale
gray = cv2.cvtColor(imcv,cv2.COLOR_RGB2GRAY)
gray = np.float32(gray)
# Find edges
edges = cv2.Canny(imcv, self.options.edge_thresh_min, self.options.edge_thresh_max, 100)
# Find coordinates of the edges
coords = [(float(x),float(y)) for y, row in enumerate(edges) for x, col in enumerate(row) if col>0]
try:
pt, idx = kmeans2(np.array(coords), self.options.num_points, minit="points")
except ValueError:
inkex.utils.debug("Too much points. Reduce sampled points and try again!")
exit(1)
if self.options.add_corners:
# Add the four corners
corners = [(0, 0),
(self.img_real_width-1, 0),
(0, self.img_real_height-1),
(self.img_real_width-1, self.img_real_height-1)]
pt = np.vstack((pt, corners))
# Perform Delaunay triangulation
tri = Delaunay(pt)
tri_coord = [(pt[t[0]], pt[t[1]], pt[t[2]]) for t in tri.simplices]
tri_colors = [(
(self.red[int(t[0][1]),int(t[0][0])], self.green[int(t[0][1]),int(t[0][0])], self.blue[int(t[0][1]),int(t[0][0])]),
(self.red[int(t[1][1]),int(t[1][0])], self.green[int(t[1][1]),int(t[1][0])], self.blue[int(t[1][1]),int(t[1][0])]),
(self.red[int(t[2][1]),int(t[2][0])], self.green[int(t[2][1]),int(t[2][0])], self.blue[int(t[2][1]),int(t[2][0])])
)
for t in tri_coord]
for i, c in enumerate(tri_coord):
# Convert to screen coordinates
v0 = self.imgToScreen(c[0][0], c[0][1])
v1 = self.imgToScreen(c[1][0], c[1][1])
v2 = self.imgToScreen(c[2][0], c[2][1])
col = tri_colors[i]
fill = ""
if self.options.gradient_fill:
color1 = "rgb("+str(col[0][0])+","+str(col[0][1])+","+str(col[0][2])+")"
color2 = "rgb("+str(0.5*col[1][0]+0.5*col[2][0])+","+ \
str(0.5*col[1][1]+0.5*col[2][1])+","+ \
str(0.5*col[1][2]+0.5*col[2][2])+")"
gradID = 'linearGradient'
# Make sure that the id is inique
index = 0
while (str(self.svg.get_ids()) in gradID):
gradID = 'linearGradient' + str(index)
index = index + 1
#self.doc_ids[gradID]=1
gradient = self.createLinearGradient(v0[0], v0[1],
0.5*(v1[0]+v2[0]), 0.5*(v1[1]+v2[1]),
color1, color2, gradID)
fill = "url(#"+gradient.get("id")+")"
else:
fill = "rgb("+str(col[0][0])+","+str(col[0][1])+","+str(col[0][2])+")"
tri_style = {
'stroke-width' : '1px',
'stroke-linecap' : 'round',
'stroke-opacity' : '1',
'fill' : fill,
'fill-opacity' : '1',
'stroke' : fill
}
self.draw_SVG_path([v0, v1, v2], 1, tri_style, grp)
if __name__ == '__main__':
ImageTriangulation().run()