# -*- coding: utf-8 -*-
import remi
import remi.gui as gui
import cv2
from threading import Timer, Thread
import traceback
import time
import math
import base64
import numpy as np
b64encoded_sample_icon = "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"
sample_icon_data = base64.b64decode(b64encoded_sample_icon)
sample_icon_data = np.fromstring(sample_icon_data, np.uint8)
sample_icon_data = cv2.imdecode(sample_icon_data, cv2.IMREAD_COLOR)
# noinspection PyUnresolvedReferences
class OpencvWidget(object):
def _setup(self):
#this must be called after the Widget super constructor
self.on_new_image.do = self.do
def do(self, callback, *userdata, **kwuserdata):
#this method gets called when an event is connected, making it possible to execute the process chain directly, before the event triggers
if hasattr(self.on_new_image.event_method_bound, '_js_code'):
self.on_new_image.event_source_instance.attributes[self.on_new_image.event_name] = self.on_new_image.event_method_bound._js_code%{
'emitter_identifier':self.on_new_image.event_source_instance.identifier, 'event_name':self.on_new_image.event_name}
self.on_new_image.callback = callback
self.on_new_image.userdata = userdata
self.on_new_image.kwuserdata = kwuserdata
#here the callback is called immediately to make it possible link to the plc
if callback is not None: #protection against the callback replacements in the editor
callback(self, *userdata, **kwuserdata)
@gui.decorate_set_on_listener("(self, emitter)")
@gui.decorate_event
def on_new_image(self, *args, **kwargs):
return ()
class OpencvImage(gui.Image, OpencvWidget):
""" OpencvImage widget.
Allows to read an image from file.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64," + b64encoded_sample_icon
app_instance = None #the application instance used to send updates
img = None #the image data as numpy array
default_style = {'position':'absolute','left':'10px','top':'10px'}
image_source = None #the linked widget instance, get updated on image listener
#OpencvImage inherits gui.Image and so it already inherits attr_src.
# I'm redefining it in order to avoid editor_attribute_decorator
# and so preventing it to be shown in editor.
@property
def attr_src(self): return self.attributes.get('src', '')
@attr_src.setter
def attr_src(self, value): self.attributes['src'] = str(value)
def __init__(self, filename='', *args, **kwargs):
self.default_style.update(kwargs.get('style',{}))
kwargs['style'] = self.default_style
kwargs['width'] = kwargs['style'].get('width', kwargs.get('width','200px'))
kwargs['height'] = kwargs['style'].get('height', kwargs.get('height','180px'))
super(OpencvImage, self).__init__(filename, *args, **kwargs)
OpencvWidget._setup(self)
def on_new_image_listener(self, emitter):
if emitter.img is None:
return
self.set_image_data(emitter.img)
def set_image(self, filename):
self.filename = filename
def set_image_data(self, img):
self.img = img
self.update()
self.on_new_image()
def search_app_instance(self, node):
if issubclass(node.__class__, remi.server.App):
return node
if not hasattr(node, "get_parent"):
return None
return self.search_app_instance(node.get_parent())
def update(self, *args):
if self.app_instance==None:
self.app_instance = self.search_app_instance(self)
if self.app_instance==None:
return
self.app_instance.execute_javascript("""
url = '/%(id)s/get_image_data?index=%(frame_index)s';
xhr = null;
xhr = new XMLHttpRequest();
xhr.open('GET', url, true);
xhr.responseType = 'blob'
xhr.onload = function(e){
urlCreator = window.URL || window.webkitURL;
urlCreator.revokeObjectURL(document.getElementById('%(id)s').src);
imageUrl = urlCreator.createObjectURL(this.response);
document.getElementById('%(id)s').src = imageUrl;
}
xhr.send();
""" % {'id': self.identifier, 'frame_index':str(time.time())})
def get_image_data(self, index=0):
gui.Image.set_image(self, '/%(id)s/get_image_data?index=%(frame_index)s'% {'id': self.identifier, 'frame_index':str(time.time())})
self._set_updated()
try:
ret, png = cv2.imencode('.png', self.img)
if ret:
headers = {'Content-type': 'image/png', 'Cache-Control':'no-cache'}
return [png.tostring(), headers]
except Exception:
pass
#print(traceback.format_exc())
return None, None
class OpencvImRead(OpencvImage, OpencvWidget):
""" OpencvImRead widget.
Allows to read an image from file.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
@property
@gui.editor_attribute_decorator("WidgetSpecific",'''Image local filename''', 'file', {})
def filename(self): return self.__filename
@filename.setter
def filename(self, value):
self.__filename = value
if len(value)>0:
self.set_image_data(cv2.imread(value, cv2.IMREAD_COLOR))
def __init__(self, filename='', *args, **kwargs):
self.default_style.update(kwargs.get('style',{}))
kwargs['style'] = self.default_style
kwargs['width'] = kwargs['style'].get('width', kwargs.get('width','200px'))
kwargs['height'] = kwargs['style'].get('height', kwargs.get('height','180px'))
super(OpencvImRead, self).__init__("", *args, **kwargs)
OpencvWidget._setup(self)
self.filename = filename
class OpencvVideo(OpencvImage):
""" OpencvVideo widget.
Opens a video source and dispatches the image frame by generating on_new_image event.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64,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"
capture = None #the cv2 VideoCapture
thread_stop_flag = False #a flag to stop the acquisition thread
thread = None # the update thread
@property
@gui.editor_attribute_decorator('WidgetSpecific','The refresh interval in Hz', int, {'default':0, 'min':0, 'max':65535, 'step':1})
def framerate(self): return self.__framerate
@framerate.setter
def framerate(self, v): self.__framerate = v;
@property
@gui.editor_attribute_decorator('WidgetSpecific','The video source index', int, {'default':0, 'min':0, 'max':65535, 'step':1})
def video_source(self): return self.__video_source
@video_source.setter
def video_source(self, v): self.__video_source = v; self.capture = cv2.VideoCapture(self.__video_source)
def __init__(self, *args, **kwargs):
self.framerate = 10
self.video_source = 0
super(OpencvVideo, self).__init__("", *args, **kwargs)
self.thread = Thread(target=self.update)
self.thread.daemon = True
self.thread.start()
def set_image_data(self, image_data_as_numpy_array):
#oveloaded to avoid update
self.img = image_data_as_numpy_array
def search_app_instance(self, node):
if issubclass(node.__class__, remi.server.App):
return node
if not hasattr(node, "get_parent"):
return None
return self.search_app_instance(node.get_parent())
def __del__(self):
self.thread_stop_flag = True
super(OpencvVideo, self).__del__()
def update(self, *args):
while not self.thread_stop_flag:
time.sleep(1.0/self.framerate)
if self.app_instance==None:
self.app_instance = self.search_app_instance(self)
if self.app_instance==None:
continue
with self.app_instance.update_lock:
self.app_instance.execute_javascript("""
var url = '/%(id)s/get_image_data?index=%(frame_index)s';
var xhr = new XMLHttpRequest();
xhr.open('GET', url, true);
xhr.responseType = 'blob'
xhr.onload = function(e){
var urlCreator = window.URL || window.webkitURL;
urlCreator.revokeObjectURL(document.getElementById('%(id)s').src);
var imageUrl = urlCreator.createObjectURL(this.response);
document.getElementById('%(id)s').src = imageUrl;
}
xhr.send();
""" % {'id': self.identifier, 'frame_index':str(time.time())})
def get_image_data(self, index=0):
gui.Image.set_image(self, '/%(id)s/get_image_data?index=%(frame_index)s'% {'id': self.identifier, 'frame_index':str(time.time())})
self._set_updated()
try:
ret, frame = self.capture.read()
if ret:
self.set_image_data(frame)
self.on_new_image()
ret, png = cv2.imencode('.png', frame)
if ret:
headers = {'Content-type': 'image/png', 'Cache-Control':'no-cache'}
# tostring is an alias to tobytes, which wasn't added till numpy 1.9
return [png.tostring(), headers]
except Exception:
print(traceback.format_exc())
return None, None
class OpencvCrop(OpencvImage):
""" OpencvCrop widget.
Allows to crop an image.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
@property
@gui.editor_attribute_decorator('WidgetSpecific','The x crop coordinate', int, {'default':0, 'min':0, 'max':65535, 'step':1})
def crop_x(self): return self.__crop_x
@crop_x.setter
def crop_x(self, v): self.__crop_x = v; self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The y crop coordinate', int, {'default':0, 'min':0, 'max':65535, 'step':1})
def crop_y(self): return self.__crop_y
@crop_y.setter
def crop_y(self, v): self.__crop_y = v; self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The width crop coordinate', int, {'default':0, 'min':0, 'max':65535, 'step':1})
def crop_w(self): return self.__crop_w
@crop_w.setter
def crop_w(self, v): self.__crop_w = v; self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The height crop coordinate', int, {'default':0, 'min':0, 'max':65535, 'step':1})
def crop_h(self): return self.__crop_h
@crop_h.setter
def crop_h(self, v): self.__crop_h = v; self.on_new_image_listener(self.image_source)
icon = "data:image/png;base64,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"
def __init__(self, *args, **kwargs):
self.crop_x = 0
self.crop_y = 0
self.crop_w = 0
self.crop_h = 0
super(OpencvCrop, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter): #CROP
if emitter is None or emitter.img is None:
return
self.image_source = emitter
self.img = emitter.img[self.crop_y:self.crop_y+self.crop_h, self.crop_x:self.crop_x+self.crop_w]
self.set_image_data(self.img)
class OpencvThreshold(OpencvImage):
""" OpencvThreshold widget.
Allows to threashold an image.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAADwAAAAuCAYAAAB04nriAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAETSURBVGhD7ZYBDsMgCEV197+zG+m60EwBHXaCvKRZslnhlT/TnFIqr2sbHu/PbdhO+BLpUnymO2fQPIhIe0ccaRwLjIW/QXekW7IA9duKqETakjQrbG2CHHFKe4cVlpzCll5YzEwYzhJ8jSISpiZ4x3RrgqPScNen4xWjSYlJ+8V7LBtpaJKb4siUlxOWiP4C7PzXSGvIcX3jGiJhrqmRB6U9RaoHXIuMNCyUNHauk6wFpOtm0BQebYq7b5asdN8phxYUrzUwS7aHqrBWY+c+rQegjaTGl7B2Y3eIYrh6UyK9Mhfhu6cxC8pj7wl7ojXlmLAnalOGb/pfhA0TkfZOCHsnhL0Twt4JYe+EsHdC2DcpPQHUiTG7/qs9SwAAAABJRU5ErkJggg=="
@property
@gui.editor_attribute_decorator('WidgetSpecific','The threshold value to binarize image', int, {'default':125, 'min':0, 'max':255, 'step':1})
def threshold(self): return self.__threshold
@threshold.setter
def threshold(self, v): self.__threshold = int(float(v)); self.on_new_image_listener(self.image_source)
def __init__(self, *args, **kwargs):
super(OpencvThreshold, self).__init__("", *args, **kwargs)
self.threshold = 125
def on_new_image_listener(self, emitter): #THRESHOLD
if emitter is None or emitter.img is None:
return
self.image_source = emitter
img = emitter.img
if len(img.shape)>2:
img = cv2.cvtColor(emitter.img, cv2.COLOR_BGR2GRAY)
res, self.img = cv2.threshold(img,self.threshold,255,cv2.THRESH_BINARY)
self.set_image_data(self.img)
'''
class OpencvSimpleBlobDetector(OpencvImage):
""" OpencvSimpleBlobDetector widget.
Allows to get blobs in an image.
Receives an image on on_new_image_listener.
The event on_blobs_detected can be connected to a listener further processing
"""
icon = "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAADwAAAAuCAYAAAB04nriAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAETSURBVGhD7ZYBDsMgCEV197+zG+m60EwBHXaCvKRZslnhlT/TnFIqr2sbHu/PbdhO+BLpUnymO2fQPIhIe0ccaRwLjIW/QXekW7IA9duKqETakjQrbG2CHHFKe4cVlpzCll5YzEwYzhJ8jSISpiZ4x3RrgqPScNen4xWjSYlJ+8V7LBtpaJKb4siUlxOWiP4C7PzXSGvIcX3jGiJhrqmRB6U9RaoHXIuMNCyUNHauk6wFpOtm0BQebYq7b5asdN8phxYUrzUwS7aHqrBWY+c+rQegjaTGl7B2Y3eIYrh6UyK9Mhfhu6cxC8pj7wl7ojXlmLAnalOGb/pfhA0TkfZOCHsnhL0Twt4JYe+EsHdC2DcpPQHUiTG7/qs9SwAAAABJRU5ErkJggg=="
def __init__(self, *args, **kwargs):
super(OpencvSimpleBlobDetector, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter): #THRESHOLD
if emitter.img is None:
return
img = emitter.img
self.set_image_data(self.img)
params = cv2.SimpleBlobDetector_Params()
params.filterByCircularity = False
params.filterByConvexity = False
params.filterByInertia = False
# I loghi appaiono di colore bianco
params.minThreshold = 100 # the graylevel of images
params.maxThreshold = 255
params.filterByColor = False
#params.blobColor = 255
# Filter by Area
params.filterByArea = True
params.minArea = 20
detector = cv2.SimpleBlobDetector_create(params) #SimpleBlobDetector()
# Detect blobs.
keypoints = detector.detect(diff_images.astype(np.uint8))
for k in keypoints:
cv2.circle(img, (int(k.pt[0]), int(k.pt[1])), 20, (255,0,0), 5)
'''
class OpencvSplit(OpencvImage):
""" OpencvSplit widget.
Splits the image channels and generates a signal for each one to dispatch the results.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
The events:
- on_new_image_first_component
- on_new_image_second_component
- on_new_image_third_component
dispatch each one a single channel.
"""
icon = "data:image/png;base64,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"
def __init__(self, *args, **kwargs):
super(OpencvSplit, self).__init__("", *args, **kwargs)
self.on_new_image_first_component.do = self.do_first
self.on_new_image_second_component.do = self.do_second
self.on_new_image_third_component.do = self.do_third
def on_new_image_listener(self, emitter):
self.image_source = emitter
self.set_image_data(emitter.img)
if not self.on_new_image_first_component.callback is None:
self.on_new_image_first_component()
if not self.on_new_image_second_component.callback is None:
self.on_new_image_second_component()
if not self.on_new_image_third_component.callback is None:
self.on_new_image_third_component()
def do_first(self, callback, *userdata, **kwuserdata):
#this method gets called when an event is connected, making it possible to execute the process chain directly, before the event triggers
if hasattr(self.on_new_image_first_component.event_method_bound, '_js_code'):
self.on_new_image_first_component.event_source_instance.attributes[self.on_new_image_first_component.event_name] = self.on_new_image_first_component.event_method_bound._js_code%{
'emitter_identifier':self.on_new_image_first_component.event_source_instance.identifier, 'event_name':self.on_new_image_first_component.event_name}
self.on_new_image_first_component.callback = callback
self.on_new_image_first_component.userdata = userdata
self.on_new_image_first_component.kwuserdata = kwuserdata
#here the callback is called immediately to make it possible link to the plc
if callback is not None: #protection against the callback replacements in the editor
if hasattr(self, "image_source"):
if not self.image_source.img is None:
self.img = cv2.split(self.image_source.img)[0]
callback(self, *userdata, **kwuserdata)
@gui.decorate_set_on_listener("(self, emitter)")
@gui.decorate_event
def on_new_image_first_component(self):
if hasattr(self, "image_source"):
if not self.image_source.img is None:
self.img = cv2.split(self.image_source.img)[0]
return ()
def do_second(self, callback, *userdata, **kwuserdata):
#this method gets called when an event is connected, making it possible to execute the process chain directly, before the event triggers
if hasattr(self.on_new_image_second_component.event_method_bound, '_js_code'):
self.on_new_image_second_component.event_source_instance.attributes[self.on_new_image_second_component.event_name] = self.on_new_image_second_component.event_method_bound._js_code%{
'emitter_identifier':self.on_new_image_second_component.event_source_instance.identifier, 'event_name':self.on_new_image_second_component.event_name}
self.on_new_image_second_component.callback = callback
self.on_new_image_second_component.userdata = userdata
self.on_new_image_second_component.kwuserdata = kwuserdata
#here the callback is called immediately to make it possible link to the plc
if callback is not None: #protection against the callback replacements in the editor
if hasattr(self, "image_source"):
if not self.image_source.img is None:
self.img = cv2.split(self.image_source.img)[1]
callback(self, *userdata, **kwuserdata)
@gui.decorate_set_on_listener("(self, emitter)")
@gui.decorate_event
def on_new_image_second_component(self):
if hasattr(self, "image_source"):
if not self.image_source.img is None:
self.img = cv2.split(self.image_source.img)[1]
return ()
def do_third(self, callback, *userdata, **kwuserdata):
#this method gets called when an event is connected, making it possible to execute the process chain directly, before the event triggers
if hasattr(self.on_new_image_third_component.event_method_bound, '_js_code'):
self.on_new_image_third_component.event_source_instance.attributes[self.on_new_image_third_component.event_name] = self.on_new_image_third_component.event_method_bound._js_code%{
'emitter_identifier':self.on_new_image_third_component.event_source_instance.identifier, 'event_name':self.on_new_image_third_component.event_name}
self.on_new_image_third_component.callback = callback
self.on_new_image_third_component.userdata = userdata
self.on_new_image_third_component.kwuserdata = kwuserdata
#here the callback is called immediately to make it possible link to the plc
if callback is not None: #protection against the callback replacements in the editor
if hasattr(self, "image_source"):
if not self.image_source.img is None:
self.img = cv2.split(self.image_source.img)[2]
callback(self, *userdata, **kwuserdata)
@gui.decorate_set_on_listener("(self, emitter)")
@gui.decorate_event
def on_new_image_third_component(self):
if hasattr(self, "image_source"):
if not self.image_source.img is None:
self.img = cv2.split(self.image_source.img)[2]
return ()
class OpencvCvtColor(OpencvImage):
""" OpencvCvtColor widget.
Convert image colorspace.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64,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"
cvt_types = {'COLOR_BGR2HSV':cv2.COLOR_BGR2HSV,'COLOR_HSV2BGR':cv2.COLOR_HSV2BGR, 'COLOR_RGB2BGR':cv2.COLOR_RGB2BGR, 'COLOR_RGB2GRAY':cv2.COLOR_RGB2GRAY, 'COLOR_BGR2GRAY':cv2.COLOR_BGR2GRAY, 'COLOR_RGB2HSV':cv2.COLOR_RGB2HSV}
@property
@gui.editor_attribute_decorator('WidgetSpecific','The conversion constant code', 'DropDown', {'possible_values': cvt_types.keys()})
def conversion_code(self):
return self.__conversion_code
@conversion_code.setter
def conversion_code(self, v):
self.__conversion_code = v
self.on_new_image_listener(self.image_source)
def __init__(self, *args, **kwargs):
self.conversion_code = cv2.COLOR_BGR2HSV
super(OpencvCvtColor, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter):
if emitter is None or emitter.img is None:
return
code = self.cvt_types[self.conversion_code] if type(self.conversion_code) == str else self.conversion_code
self.set_image_data(cv2.cvtColor(emitter.img, code))
class OpencvBitwiseNot(OpencvImage):
""" OpencvBitwiseNot widget.
Allows to invert an image mask.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64,"+base64.b64encode( cv2.imencode('.png', cv2.bitwise_not(cv2.threshold(cv2.cvtColor(sample_icon_data, cv2.COLOR_BGR2GRAY),130,255,cv2.THRESH_BINARY)[1]) )[1] ).decode("utf-8")
def __init__(self, *args, **kwargs):
super(OpencvBitwiseNot, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter):
try:
self.set_image_data(cv2.bitwise_not(emitter.img))
except Exception:
print(traceback.format_exc())
class BinaryOperator(object):
img1 = None
img2 = None
def process(self):
#overload this method to perform different operations
if not self.img1 is None:
if not self.img2 is None:
pass
def on_new_image_1_listener(self, emitter):
try:
self.img1 = emitter.img
self.process()
except Exception:
print(traceback.format_exc())
def on_new_image_2_listener(self, emitter):
try:
self.img2 = emitter.img
self.process()
except Exception:
print(traceback.format_exc())
class OpencvBitwiseAnd(OpencvImage, BinaryOperator):
""" OpencvBitwiseAnd widget.
Allows to do the AND of two images.
- Receives the image on on_new_image_1_listener.
- Receives the mask on on_new_image_2_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = None
def __init__(self, *args, **kwargs):
BinaryOperator.__init__(self)
super(OpencvBitwiseAnd, self).__init__("", *args, **kwargs)
def process(self):
if not self.img1 is None:
if not self.img2 is None:
self.set_image_data(cv2.bitwise_and(self.img1, self.img1, mask=self.img2))
class OpencvBitwiseOr(OpencvImage, BinaryOperator):
""" OpencvBitwiseOr widget.
Allows to do the OR of two images.
- Receives the image on on_new_image_1_listener.
- Receives the mask on on_new_image_2_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = None
def __init__(self, *args, **kwargs):
BinaryOperator.__init__(self)
super(OpencvBitwiseOr, self).__init__("", *args, **kwargs)
def process(self):
if not self.img1 is None:
if not self.img2 is None:
self.set_image_data(cv2.bitwise_or(self.img1, self.img1, mask=self.img2))
class OpencvAddWeighted(OpencvImage, BinaryOperator):
""" OpencvAddWeighted widget.
Allows to do the add_weighted of two images.
- Receives first image on on_new_image_1_listener.
- Receives second mask on on_new_image_2_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64,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"
@property
@gui.editor_attribute_decorator('WidgetSpecific','The alpha value', float, {'default':0, 'min':0, 'max':1.0, 'step':0.0001})
def alpha(self): return self.__alpha
@alpha.setter
def alpha(self, v): self.__alpha = v; self.process()
@property
@gui.editor_attribute_decorator('WidgetSpecific','The beta value', float, {'default':0, 'min':0, 'max':1.0, 'step':0.0001})
def beta(self): return self.__beta
@beta.setter
def beta(self, v): self.__beta = v; self.process()
@property
@gui.editor_attribute_decorator('WidgetSpecific','The gamma value', float, {'default':0, 'min':0, 'max':1.0, 'step':0.0001})
def gamma(self): return self.__gamma
@gamma.setter
def gamma(self, v): self.__gamma = v; self.process()
def __init__(self, *args, **kwargs):
self.alpha = 0.5
self.beta = 0.5
self.gamma = 0.0
BinaryOperator.__init__(self)
super(OpencvAddWeighted, self).__init__("", *args, **kwargs)
def process(self):
if not self.img1 is None:
if not self.img2 is None:
self.set_image_data(cv2.addWeighted(self.img1, self.__alpha, self.img2, self.__beta, self.__gamma))
class OpencvBilateralFilter(OpencvImage):
""" OpencvBilateralFilter widget.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
border_type = {"BORDER_CONSTANT": cv2.BORDER_CONSTANT,
"BORDER_REPLICATE": cv2.BORDER_REPLICATE,
"BORDER_REFLECT": cv2.BORDER_REFLECT,
"BORDER_WRAP": cv2.BORDER_WRAP,
"BORDER_REFLECT_101": cv2.BORDER_REFLECT_101,
"BORDER_TRANSPARENT": cv2.BORDER_TRANSPARENT,
"BORDER_REFLECT101": cv2.BORDER_REFLECT101,
"BORDER_DEFAULT": cv2.BORDER_DEFAULT,
"BORDER_ISOLATED": cv2.BORDER_ISOLATED}
icon = "data:image/png;base64,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"
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter diameter', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def diameter(self):
return self.__diameter
@diameter.setter
def diameter(self, v):
self.__diameter = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter sigma color parameter', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def sigma_color(self):
return self.__sigma_color
@sigma_color.setter
def sigma_color(self, v):
self.__sigma_color = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter sigma space parameter', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def sigma_space(self):
return self.__sigma_space
@sigma_space.setter
def sigma_space(self, v):
self.__sigma_space = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter border parameter', 'DropDown', {'possible_values': border_type.keys()})
def border(self):
return self.__border
@border.setter
def border(self, v):
self.__border = v
self.on_new_image_listener(self.image_source)
def __init__(self, diameter=2, sigma_color=0, sigma_space=0, border=cv2.BORDER_CONSTANT, *args, **kwargs):
self.__sigma_color = sigma_color
self.__sigma_space = sigma_space
self.__diameter = diameter
self.__border = border
super(OpencvBilateralFilter, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter):
try:
self.image_source = emitter
border = self.border_type[self.border] if type(self.border) == str else self.border
self.set_image_data(cv2.bilateralFilter(emitter.img, self.diameter, self.sigma_color, self.sigma_space, borderType=border))
except Exception:
print(traceback.format_exc())
class OpencvBlurFilter(OpencvImage):
""" OpencvBlurFilter widget.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64,"+base64.b64encode( cv2.imencode('.png', cv2.blur(sample_icon_data,(6,6)) )[1] ).decode("utf-8")
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter kernel_size', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def kernel_size(self):
return self.__kernel_size
@kernel_size.setter
def kernel_size(self, v):
self.__kernel_size = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter border parameter', 'DropDown', {'possible_values': OpencvBilateralFilter.border_type.keys()})
def border(self):
return self.__border
@border.setter
def border(self, v):
self.__border = v
self.on_new_image_listener(self.image_source)
def __init__(self, kernel_size=2, border=cv2.BORDER_CONSTANT, *args, **kwargs):
self.__kernel_size = kernel_size
self.__border = border
super(OpencvBlurFilter, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter):
try:
self.image_source = emitter
border = OpencvBilateralFilter.border_type[self.border] if type(self.border) == str else self.border
self.set_image_data(cv2.blur(emitter.img, (self.kernel_size,self.kernel_size), borderType=border))
except Exception:
print(traceback.format_exc())
def on_kernel_size_listener(self, emitter, value=None):
v = emitter.get_value() if value is None else value
v = int(v)
self.kernel_size = v
if hasattr(self, "image_source"):
self.on_new_image_listener(self.image_source)
class OpencvDilateFilter(OpencvImage):
""" OpencvDilateFilter widget.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
morph_shape = {"MORPH_RECT": cv2.MORPH_RECT, "MORPH_CROSS": cv2.MORPH_CROSS, "MORPH_ELLIPSE": cv2.MORPH_ELLIPSE}
icon = "data:image/png;base64,"+base64.b64encode( cv2.imencode('.png', cv2.dilate(cv2.threshold(cv2.cvtColor(sample_icon_data, cv2.COLOR_BGR2GRAY),130,255,cv2.THRESH_BINARY)[1], cv2.getStructuringElement(cv2.MORPH_RECT, (6, 6)), iterations=1) )[1] ).decode("utf-8")
@property
@gui.editor_attribute_decorator('WidgetSpecific','The kernel morph shape', 'DropDown', {'possible_values': morph_shape.keys()})
def kernel_morph_shape(self):
return self.__kernel_morph_shape
@kernel_morph_shape.setter
def kernel_morph_shape(self, v):
self.__kernel_morph_shape = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter kernel_size', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def kernel_size(self):
return self.__kernel_size
@kernel_size.setter
def kernel_size(self, v):
self.__kernel_size = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter iterations', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def iterations(self):
return self.__iterations
@iterations.setter
def iterations(self, v):
self.__iterations = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter border parameter', 'DropDown', {'possible_values': OpencvBilateralFilter.border_type.keys()})
def border(self):
return self.__border
@border.setter
def border(self, v):
self.__border = v
self.on_new_image_listener(self.image_source)
def __init__(self, kernel_morph_shape=cv2.MORPH_RECT, kernel_size=2, iterations=1, border=cv2.BORDER_CONSTANT, *args, **kwargs):
self.__kernel_morph_shape = kernel_morph_shape
self.__kernel_size = kernel_size
self.__iterations = iterations
self.__border = border
super(OpencvDilateFilter, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter):
try:
self.image_source = emitter
_kernel_morph_shape = self.morph_shape[self.kernel_morph_shape] if type(self.kernel_morph_shape) == str else self.kernel_morph_shape
kernel = cv2.getStructuringElement(_kernel_morph_shape, (self.kernel_size, self.kernel_size))
border = OpencvBilateralFilter.border_type[self.border] if type(self.border) == str else self.border
self.set_image_data(cv2.dilate(emitter.img, kernel, iterations=self.iterations, borderType=border))
except Exception:
print(traceback.format_exc())
class OpencvErodeFilter(OpencvDilateFilter):
""" OpencvErodeFilter widget.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64,"+base64.b64encode( cv2.imencode('.png', cv2.erode(cv2.threshold(cv2.cvtColor(sample_icon_data, cv2.COLOR_BGR2GRAY),130,255,cv2.THRESH_BINARY)[1], cv2.getStructuringElement(cv2.MORPH_RECT, (6, 6)), iterations=1) )[1] ).decode("utf-8")
def on_new_image_listener(self, emitter):
try:
self.image_source = emitter
_kernel_morph_shape = self.morph_shape[self.kernel_morph_shape] if type(self.kernel_morph_shape) == str else self.kernel_morph_shape
kernel = cv2.getStructuringElement(_kernel_morph_shape, (self.kernel_size, self.kernel_size))
border = OpencvBilateralFilter.border_type[self.border] if type(self.border) == str else self.border
self.set_image_data(cv2.erode(emitter.img, kernel, iterations=self.iterations, borderType=border))
except Exception:
print(traceback.format_exc())
class OpencvLaplacianFilter(OpencvImage):
""" OpencvLaplacianFilter widget.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64,"+base64.b64encode( cv2.imencode('.png', cv2.Laplacian(cv2.cvtColor(sample_icon_data, cv2.COLOR_BGR2GRAY), -1) )[1] ).decode("utf-8")
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter border parameter', 'DropDown', {'possible_values': OpencvBilateralFilter.border_type.keys()})
def border(self):
return self.__border
@border.setter
def border(self, v):
self.__border = v
self.on_new_image_listener(self.image_source)
def __init__(self, border=cv2.BORDER_CONSTANT, *args, **kwargs):
self.__border = border
super(OpencvLaplacianFilter, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter):
try:
self.image_source = emitter
border = OpencvBilateralFilter.border_type[self.border] if type(self.border) == str else self.border
self.set_image_data(cv2.Laplacian(emitter.img, -1, borderType=border))
except Exception:
print(traceback.format_exc())
class OpencvCanny(OpencvImage):
""" OpencvCanny segmentation widget.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = "data:image/png;base64,"+base64.b64encode( cv2.imencode('.png', cv2.Canny(cv2.cvtColor(sample_icon_data, cv2.COLOR_BGR2GRAY), 50, 130) )[1] ).decode("utf-8")
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter threshold1', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def threshold1(self):
return self.__threshold1
@threshold1.setter
def threshold1(self, v):
self.__threshold1 = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter threshold2', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def threshold2(self):
return self.__threshold2
@threshold2.setter
def threshold2(self, v):
self.__threshold2 = v
self.on_new_image_listener(self.image_source)
def __init__(self, threshold1=80, threshold2=160, *args, **kwargs):
self.__threshold1 = threshold1
self.__threshold2 = threshold2
super(OpencvCanny, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter):
try:
self.image_source = emitter
self.set_image_data(cv2.Canny(emitter.img, self.threshold1, self.threshold2))
except Exception:
print(traceback.format_exc())
def on_threshold1_listener(self, emitter, value=None):
v = emitter.get_value() if value is None else value
v = int(v)
self.threshold1 = v
if hasattr(self, "image_source"):
self.on_new_image_listener(self.image_source)
def on_threshold2_listener(self, emitter, value=None):
v = emitter.get_value() if value is None else value
v = int(v)
self.threshold2 = v
if hasattr(self, "image_source"):
self.on_new_image_listener(self.image_source)
#https://docs.opencv.org/3.4/d3/dc0/group__imgproc__shape.html
class OpencvFindContours(OpencvImage):
""" OpencvFindContours segmentation widget.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = None
contours = None #the contours result of processing
hierarchy = None #the hierarchy result of processing
contour_retrieval_mode = {"RETR_LIST": cv2.RETR_LIST, "RETR_EXTERNAL": cv2.RETR_EXTERNAL, "RETR_CCOMP ": cv2.RETR_CCOMP, "RETR_TREE": cv2.RETR_TREE, "RETR_FLOODFILL": cv2.RETR_FLOODFILL}
contour_approximation_method = {"CHAIN_APPROX_NONE":cv2.CHAIN_APPROX_NONE, "CHAIN_APPROX_SIMPLE": cv2.CHAIN_APPROX_SIMPLE, "CHAIN_APPROX_TC89_L1": cv2.CHAIN_APPROX_TC89_L1, "CHAIN_APPROX_TC89_KCOS": cv2.CHAIN_APPROX_TC89_KCOS}
@property
@gui.editor_attribute_decorator('WidgetSpecific','The contour retrieval mode parameter', 'DropDown', {'possible_values': contour_retrieval_mode.keys()})
def retrieval_mode(self):
return self.__retrieval_mode
@retrieval_mode.setter
def retrieval_mode(self, v):
self.__retrieval_mode = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The contour approximation method parameter', 'DropDown', {'possible_values': contour_approximation_method.keys()})
def approximation_method(self):
return self.__approximation_method
@approximation_method.setter
def approximation_method(self, v):
self.__approximation_method = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The minimum arc length of a contour', int, {'possible_values': '', 'min': 0, 'max': 9223372036854775807, 'default': 1, 'step': 1})
def min_arc_length(self):
return self.__min_arc_length
@min_arc_length.setter
def min_arc_length(self, v):
self.__min_arc_length = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The minimum arc length of a contour', int, {'possible_values': '', 'min': 0, 'max': 9223372036854775807, 'default': 1, 'step': 1})
def max_arc_length(self):
return self.__max_arc_length
@max_arc_length.setter
def max_arc_length(self, v):
self.__max_arc_length = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The minimum contour area', int, {'possible_values': '', 'min': 0, 'max': 9223372036854775807, 'default': 1, 'step': 1})
def min_contour_area(self):
return self.__min_contour_area
@min_contour_area.setter
def min_contour_area(self, v):
self.__min_contour_area = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The maximum contour area', int, {'possible_values': '', 'min': 0, 'max': 9223372036854775807, 'default': 1, 'step': 1})
def max_contour_area(self):
return self.__max_contour_area
@max_contour_area.setter
def max_contour_area(self, v):
self.__max_contour_area = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The maximum contour area', int, {'possible_values': '', 'min': 0, 'max': 9223372036854775807, 'default': 1, 'step': 1})
def max_arc_length(self):
return self.__max_arc_length
@max_arc_length.setter
def max_arc_length(self, v):
self.__max_arc_length = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','When true, convex contours are discarded', bool, {})
def discard_convex(self):
return self.__discard_convex
@discard_convex.setter
def discard_convex(self, v):
self.__discard_convex = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','When true, non-convex contours are discarded', bool, {})
def discard_non_convex(self):
return self.__discard_non_convex
@discard_non_convex.setter
def discard_non_convex(self, v):
self.__discard_non_convex = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The minimum acceptable circularity', float, {'possible_values': '', 'min': 0.0, 'max': 1.0, 'default': 0.0, 'step': 0.01})
def min_roundness(self):
return self.__min_roundness
@min_roundness.setter
def min_roundness(self, v):
self.__min_roundness = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The maximum acceptable circularity', float, {'possible_values': '', 'min': 0.0, 'max': 1.0, 'default': 0.0, 'step': 0.01})
def max_roundness(self):
return self.__max_roundness
@max_roundness.setter
def max_roundness(self, v):
self.__max_roundness = v
self.on_new_image_listener(self.image_source)
def __init__(self, retrieval_mode=cv2.RETR_LIST, approximation_method=cv2.CHAIN_APPROX_SIMPLE, *args, **kwargs):
self.__retrieval_mode = retrieval_mode
self.__approximation_method = approximation_method
self.__min_arc_length = 0
self.__max_arc_length = 9223372036854775807
self.__min_contour_area = 0
self.__max_contour_area = 9223372036854775807
self.__discard_convex = False
self.__discard_non_convex = False
self.__min_roundness = 0.0
self.__max_roundness = 1.0
super(OpencvFindContours, self).__init__("", *args, **kwargs)
self.on_new_contours_result.do = self.do_contours_result
def on_new_image_listener(self, emitter):
try:
self.image_source = emitter
if emitter.img is None:
return
_retrieval_mode = self.contour_retrieval_mode[self.retrieval_mode] if type(self.retrieval_mode) == str else self.retrieval_mode
_approximation_method = self.contour_approximation_method[self.approximation_method] if type(self.approximation_method) == str else self.approximation_method
major = cv2.__version__.split('.')[0]
img = emitter.img.copy()
if major == '3':
img, self.contours, self.hierarchy = cv2.findContours(img, _retrieval_mode, _approximation_method)
else:
self.contours, self.hierarchy = cv2.findContours(img, _retrieval_mode, _approximation_method)
filtered_contours_indices = []
for ic in range(0, len(self.contours)):
c = self.contours[ic]
if not (self.__discard_convex and cv2.isContourConvex(c)):
if not (self.__discard_non_convex and not cv2.isContourConvex(c)):
l = cv2.arcLength(c, True)
if l>self.__min_arc_length and l<self.__max_arc_length:
area = cv2.contourArea(c)
if area>self.__min_contour_area and area<self.__max_contour_area:
#https://answers.opencv.org/question/21101/circularity-of-a-connected-component/
roundness = (4.0*area) / (math.pi* (l/math.pi)**2) #4 Area / (pi Max-diam^2)
if roundness>self.__min_roundness and roundness<self.__max_roundness:
filtered_contours_indices.append(ic)
#drawing selected contours
img.fill(255)
for i in filtered_contours_indices:
img = cv2.drawContours(img, self.contours, i, 0, 1, cv2.LINE_AA)
self.set_image_data(img)
self.on_new_contours_result()
except Exception:
print(traceback.format_exc())
def do_contours_result(self, callback, *userdata, **kwuserdata):
#this method gets called when an event is connected, making it possible to execute the process chain directly, before the event triggers
if hasattr(self.on_new_contours_result.event_method_bound, '_js_code'):
self.on_new_contours_result.event_source_instance.attributes[self.on_new_contours_result.event_name] = self.on_new_contours_result.event_method_bound._js_code%{
'emitter_identifier':self.on_new_contours_result.event_source_instance.identifier, 'event_name':self.on_new_contours_result.event_name}
self.on_new_contours_result.callback = callback
self.on_new_contours_result.userdata = userdata
self.on_new_contours_result.kwuserdata = kwuserdata
#here the callback is called immediately to make it possible link to the plc
if callback is not None: #protection against the callback replacements in the editor
callback(self, self.contours, self.hierarchy, *userdata, **kwuserdata)
@gui.decorate_set_on_listener("(self, emitter, contours, hierarchy)")
@gui.decorate_event
def on_new_contours_result(self):
return (self.contours, self.hierarchy)
class OpencvInRangeGrayscale(OpencvImage):
""" OpencvInRangeGrayscale thresholding widget.
Receives an image on on_new_image_listener.
The event on_new_image can be connected to other Opencv widgets for further processing
"""
icon = None
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter threshold1', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def threshold1(self):
return self.__threshold1
@threshold1.setter
def threshold1(self, v):
self.__threshold1 = v
self.on_new_image_listener(self.image_source)
@property
@gui.editor_attribute_decorator('WidgetSpecific','The filter threshold2', int, {'possible_values': '', 'min': 0, 'max': 65535, 'default': 1, 'step': 1})
def threshold2(self):
return self.__threshold2
@threshold2.setter
def threshold2(self, v):
self.__threshold2 = v
self.on_new_image_listener(self.image_source)
def __init__(self, threshold1=80, threshold2=160, *args, **kwargs):
self.__threshold1 = threshold1
self.__threshold2 = threshold2
super(OpencvInRangeGrayscale, self).__init__("", *args, **kwargs)
def on_new_image_listener(self, emitter):
try:
self.image_source = emitter
self.set_image_data(cv2.inRange(emitter.img, self.threshold1, self.threshold2))
except Exception:
print(traceback.format_exc())
