def _updateOpenRecentMenu(self):
self.m_open_recent.clear()
for s in self.app.session.app_opts['recent sessions']:
s = PathStr(s)
a = self.m_open_recent.addAction(s.basename())
a.setToolTip(s)
a.triggered.connect(lambda checked, s=s: self.app.session.new(s))
def equalizeImage(img, save_path=None, name_additive='_eqHist'):
'''
Equalize the histogram (contrast) of an image
works with RGB/multi-channel images
and flat-arrays
@param img - image_path or np.array
@param save_path if given output images will be saved there
@param name_additive if given this additive will be appended to output images
@return output images if input images are numpy.arrays and no save_path is given
@return None elsewise
'''
if isinstance(img, basestring):
img = PathStr(img)
if not img.exists():
raise Exception("image path doesn't exist")
img_name = img.basename().replace('.','%s.' %name_additive)
if save_path is None:
save_path = img.dirname()
img = cv2.imread(img)
if img.dtype != np.dtype('uint8'):
#openCV cannot work with float arrays or uint > 8bit
eqFn = _equalizeHistogram
else:
eqFn = cv2.equalizeHist
if len(img.shape) == 3:#multi channel img like rgb
for i in range(img.shape[2]):
img[:, :, i] = eqFn(img[:, :, i])
else: # grey scale image
img = eqFn(img)
if save_path:
img_name = PathStr(save_path).join(img_name)
cv2.imwrite(img_name, img)
return img
def equalizeImage(img, save_path=None, name_additive='_eqHist'):
'''
Equalize the histogram (contrast) of an image
works with RGB/multi-channel images
and flat-arrays
@param img - image_path or np.array
@param save_path if given output images will be saved there
@param name_additive if given this additive will be appended to output images
@return output images if input images are numpy.arrays and no save_path is given
@return None elsewise
'''
if isinstance(img, string_types):
img = PathStr(img)
if not img.exists():
raise Exception("image path doesn't exist")
img_name = img.basename().replace('.', '%s.' % name_additive)
if save_path is None:
save_path = img.dirname()
img = cv2.imread(img)
if img.dtype != np.dtype('uint8'):
# openCV cannot work with float arrays or uint > 8bit
eqFn = _equalizeHistogram
else:
eqFn = cv2.equalizeHist
if len(img.shape) == 3: # multi channel img like rgb
for i in range(img.shape[2]):
img[:, :, i] = eqFn(img[:, :, i])
else: # grey scale image
img = eqFn(img)
if save_path:
img_name = PathStr(save_path).join(img_name)
cv2.imwrite(img_name, img)
return img
