def setParams(self, i_resize_scale=1, i_filter_size=0, i_eq=False, i_roi=None):
self.__resize_scale = i_resize_scale
self.__filter_size = i_filter_size
if not i_roi == None:
self.__roi = cv.cvRect( i_roi.x, i_roi.y, i_roi.width, i_roi.height )
else:
self.__roi = None
self.__equalise_hist = i_eq
def Numpy2CvRect(i_min_row=None, i_min_col=None, i_max_row=None, i_max_col=None, i_face_roi=None):
"""Convert roi from matrix format (min_row, min_col, max_row, max_col) to opencv rect"""
if i_face_roi is not None:
(i_min_row, i_min_col, i_max_row, i_max_col) = i_face_roi
y = i_min_row
x = i_min_col
height = i_max_row - y
width =i_max_col - x
return cv.cvRect(x, y, width, height)
def jitter( self, i_image , i_n_examples ):
"""1) Apply various random affine transform to i_image (scale, rotation),
where i_n_examples is the number of transformations. The affine transforms happen
around the center of the original region of interest.
2) Translate roi with various values - crop the image from this region.
3) The same normalisation is then applied to all the cropped images, specified by setParams"""
#Store transforms applied in format tx ty scale angle
o_transforms = numpy.array([0., 0., 1., 0.])
if self.__roi == None:
print "No region of interest - returning input image!"
return None
#Always return the normalised verion of the input image
image = cv.Ipl2NumPy(self.normalise(i_image))
o_data = numpy.tile( None, (image.shape[0], image.shape[1], 1))
o_data[:,:,0] = image
if i_n_examples == 0:
return ( o_data, o_transforms)
#Rotation point should be around original roi center
center = cv.cvPoint2D32f( self.__roi.x + self.__roi.width/2,self.__roi.y + self.__roi.height/2 )
angles = numpy.random.uniform(-30.,30.,i_n_examples)
scales = numpy.random.uniform(0.9,1.1, i_n_examples)
tx = numpy.int32( numpy.round( numpy.random.uniform(-20,20,i_n_examples)))
ty = numpy.int32( numpy.round( numpy.random.uniform(-20,20,i_n_examples)))
x = self.__roi.x + tx
y = self.__roi.y + ty
#FIXME: Extend valid indices to outliers due to affine transform!!!
min_x = 0; min_y = 0;
max_x = i_image.width - self.__roi.width - 1
max_y = i_image.height - self.__roi.height - 1
valid_x = numpy.hstack([numpy.nonzero( x >= min_x )[0], numpy.nonzero( x < max_x)[0]])
valid_y = numpy.hstack([numpy.nonzero( y >= min_y )[0], numpy.nonzero( y < max_y)[0]])
valid_idx = numpy.unique(numpy.hstack([valid_x, valid_y]))
original_roi = cv.cvRect( self.__roi.x, self.__roi.y, self.__roi.width, self.__roi.height)
if self.__rot_mat == None:
original_rot_matrix = None
else:
original_rot_matrix = cv.cvCloneImage(self.__rot_mat)
for index in valid_idx:
params = numpy.array([tx[index], ty[index], scales[index], angles[index] ])
self.setAffineTransform( center, scales[index], angles[index])
self.__roi.x = int( x[index] )
self.__roi.y = int( y[index] )
image = cv.Ipl2NumPy(self.normalise( i_image ))
o_data = numpy.dstack( [o_data, image])
o_transforms = numpy.vstack( [o_transforms, params])
#Restore the original region of interest
self.__roi.x = original_roi.x
self.__roi.y = original_roi.y
if original_rot_matrix == None:
self.__rot_mat= None
else:
self.__rot_mat = cv.cvCloneImage(original_rot_matrix)
return (o_data, o_transforms)
def setParams(self,
i_resize_scale=1,
i_filter_size=0,
i_eq=False,
i_roi=None):
self.__resize_scale = i_resize_scale
self.__filter_size = i_filter_size
if not i_roi == None:
self.__roi = cv.cvRect(i_roi.x, i_roi.y, i_roi.width, i_roi.height)
else:
self.__roi = None
self.__equalise_hist = i_eq
def similarityTransform(self, i_image, i_transform, i_crop=True ):
"""Apply affine transform around center of region of interest, then translate roi"""
tx = numpy.int32(numpy.round(i_transform[0]))
ty = numpy.int32(numpy.round(i_transform[1]))
scale = i_transform[2]
angle = i_transform[3]
center = cv.cvPoint2D32f( self.__roi.x + self.__roi.width/2,self.__roi.y + self.__roi.height/2 )
self.setAffineTransform( center, scale, angle)
original_roi = cv.cvRect( self.__roi.x, self.__roi.y, self.__roi.width, self.__roi.height)
if not i_crop:
self.__roi = None
else:
self.__roi.x = self.__roi.x + int(tx)
self.__roi.y = self.__roi.y + int(ty)
filter_size = self.__filter_size
self.__filter_size = 0
o_image = self.normalise(i_image)
self.__filter_size = filter_size
self.__roi = cv.cvRect(original_roi.x, original_roi.y, original_roi.width, original_roi.height)
return cv.Ipl2NumPy(o_image)
def Numpy2CvRect(i_min_row=None,
i_min_col=None,
i_max_row=None,
i_max_col=None,
i_face_roi=None):
"""Convert roi from matrix format (min_row, min_col, max_row, max_col) to opencv rect"""
if i_face_roi is not None:
(i_min_row, i_min_col, i_max_row, i_max_col) = i_face_roi
y = i_min_row
x = i_min_col
height = i_max_row - y
width = i_max_col - x
return cv.cvRect(x, y, width, height)
def detect(self, camera):
# Kamera veya görüntüden bir kare al
frame = cvQueryFrame(camera)
if not frame:
return (None, None)
cropped = cvCreateImage(cvSize(240, 240), 8, 3)
src_region = cvGetSubRect(frame, opencv.cvRect(0, 0, 240, 240))
cvCopy(src_region, cropped)
frame = cropped
# cvQueryFrame ile alınan görüntü değiştirilemeyeceğinden, kopyala
copy = cvCreateImage(cvSize(frame.width, frame.height), IPL_DEPTH_8U,
frame.nChannels)
# Eğer kameradan alınan görüntünün orijin noktası sol üst köşe ise(top down), kopyala
if (frame.origin == IPL_ORIGIN_TL):
cvCopy(frame, copy)
# Değilse(bottom up IPL_ORIGIN_BL ise), x ekseni etrafında döndür ve kopyala
else:
cvFlip(frame, copy, 0)
# image_scale a göre küçült
small_img = cvCreateImage(
cvSize(cvRound(copy.width / self.image_scale),
cvRound(copy.height / self.image_scale)), 8, 1)
# küçültülmemiş kopyasını yap
gray = cvCreateImage(cvSize(copy.width, copy.height), 8, 1)
# grileştir
cvCvtColor(copy, gray, CV_BGR2GRAY)
# küçük gri halini small_img'a koy
cvResize(gray, small_img, CV_INTER_LINEAR)
# parlaklığını normalleştir ve kontrastı arttır
cvEqualizeHist(small_img, small_img)
# storage'ı temizle. bu fonksiyon hafızadan alanı deallocate etmez!
cvClearMemStorage(self.storage)
# Yüz tanıma işlemi
t = cvGetTickCount()
faces = cvHaarDetectObjects(
small_img, # yüzlerin aranacağı görüntü
self.cascade, # önceden belirli tanıma verisi
self.storage, # sonuçların tutulacağı hafıza bölgesi
# aşağıdakilerin açıklaması sınıf tanımlamasının başında
self.haar_scale,
self.min_neighbors,
self.haar_flags,
self.min_size)
t = cvGetTickCount() - t
#print "detection time = %gms" % (t/(cvGetTickFrequency()*1000.))
return frame, faces
def similarityTransform(self, i_image, i_transform, i_crop=True):
"""Apply affine transform around center of region of interest, then translate roi"""
tx = numpy.int32(numpy.round(i_transform[0]))
ty = numpy.int32(numpy.round(i_transform[1]))
scale = i_transform[2]
angle = i_transform[3]
center = cv.cvPoint2D32f(self.__roi.x + self.__roi.width / 2,
self.__roi.y + self.__roi.height / 2)
self.setAffineTransform(center, scale, angle)
original_roi = cv.cvRect(self.__roi.x, self.__roi.y, self.__roi.width,
self.__roi.height)
if not i_crop:
self.__roi = None
else:
self.__roi.x = self.__roi.x + int(tx)
self.__roi.y = self.__roi.y + int(ty)
filter_size = self.__filter_size
self.__filter_size = 0
o_image = self.normalise(i_image)
self.__filter_size = filter_size
self.__roi = cv.cvRect(original_roi.x, original_roi.y,
original_roi.width, original_roi.height)
return cv.Ipl2NumPy(o_image)
def detect(self,camera):
# Kamera veya görüntüden bir kare al
frame = cvQueryFrame(camera)
if not frame:
return (None,None)
cropped = cvCreateImage( cvSize(240, 240), 8, 3)
src_region = cvGetSubRect(frame, opencv.cvRect(0, 0, 240, 240) )
cvCopy(src_region, cropped)
frame = cropped
# cvQueryFrame ile alınan görüntü değiştirilemeyeceğinden, kopyala
copy = cvCreateImage(cvSize(frame.width, frame.height), IPL_DEPTH_8U, frame.nChannels)
# Eğer kameradan alınan görüntünün orijin noktası sol üst köşe ise(top down), kopyala
if (frame.origin == IPL_ORIGIN_TL):
cvCopy(frame, copy)
# Değilse(bottom up IPL_ORIGIN_BL ise), x ekseni etrafında döndür ve kopyala
else:
cvFlip(frame, copy, 0)
# image_scale a göre küçült
small_img = cvCreateImage(cvSize(cvRound(copy.width / self.image_scale), cvRound(copy.height / self.image_scale)), 8, 1)
# küçültülmemiş kopyasını yap
gray = cvCreateImage(cvSize(copy.width, copy.height), 8, 1)
# grileştir
cvCvtColor(copy, gray, CV_BGR2GRAY)
# küçük gri halini small_img'a koy
cvResize(gray, small_img, CV_INTER_LINEAR)
# parlaklığını normalleştir ve kontrastı arttır
cvEqualizeHist(small_img, small_img)
# storage'ı temizle. bu fonksiyon hafızadan alanı deallocate etmez!
cvClearMemStorage(self.storage)
# Yüz tanıma işlemi
t = cvGetTickCount()
faces = cvHaarDetectObjects(
small_img, # yüzlerin aranacağı görüntü
self.cascade, # önceden belirli tanıma verisi
self.storage, # sonuçların tutulacağı hafıza bölgesi
# aşağıdakilerin açıklaması sınıf tanımlamasının başında
self.haar_scale,
self.min_neighbors,
self.haar_flags,
self.min_size
)
t = cvGetTickCount() - t
#print "detection time = %gms" % (t/(cvGetTickFrequency()*1000.))
return frame, faces
def tile_images(img_width, img_height, num_width, num_height, images, channels=3):
w = img_width * num_width
h = img_height * num_height
image = cv.cvCreateImage(cv.cvSize(int(w), int(h)), 8, channels)
cv.cvSet(image, cv.cvScalar(255,255,255))
while len(images) > 0:
try:
for y in range(int(num_height)):
for x in range(int(num_width)):
small_tile = images.pop()
img_x = x * img_width
img_y = y * img_height
cropped = cv.cvGetSubRect(image, cv.cvRect(img_x, img_y, img_width,img_height))
cv.cvCopy(small_tile, cropped)
except exceptions.IndexError, e:
break
def tile_images(img_width,
img_height,
num_width,
num_height,
images,
channels=3):
w = img_width * num_width
h = img_height * num_height
image = cv.cvCreateImage(cv.cvSize(int(w), int(h)), 8, channels)
cv.cvSet(image, cv.cvScalar(255, 255, 255))
while len(images) > 0:
try:
for y in range(int(num_height)):
for x in range(int(num_width)):
small_tile = images.pop()
img_x = x * img_width
img_y = y * img_height
cropped = cv.cvGetSubRect(
image, cv.cvRect(img_x, img_y, img_width, img_height))
cv.cvCopy(small_tile, cropped)
except exceptions.IndexError, e:
break
def setRoi(self, i_roi):
self.__roi = cv.cvRect(i_roi.x, i_roi.y, i_roi.width, i_roi.height)
def getRoi(self):
if self.__roi is None:
return None
return cv.cvRect(self.__roi.x, self.__roi.y, self.__roi.width,
self.__roi.height)
def jitter(self, i_image, i_n_examples):
"""1) Apply various random affine transform to i_image (scale, rotation),
where i_n_examples is the number of transformations. The affine transforms happen
around the center of the original region of interest.
2) Translate roi with various values - crop the image from this region.
3) The same normalisation is then applied to all the cropped images, specified by setParams"""
#Store transforms applied in format tx ty scale angle
o_transforms = numpy.array([0., 0., 1., 0.])
if self.__roi == None:
print "No region of interest - returning input image!"
return None
#Always return the normalised verion of the input image
image = cv.Ipl2NumPy(self.normalise(i_image))
o_data = numpy.tile(None, (image.shape[0], image.shape[1], 1))
o_data[:, :, 0] = image
if i_n_examples == 0:
return (o_data, o_transforms)
#Rotation point should be around original roi center
center = cv.cvPoint2D32f(self.__roi.x + self.__roi.width / 2,
self.__roi.y + self.__roi.height / 2)
angles = numpy.random.uniform(-30., 30., i_n_examples)
scales = numpy.random.uniform(0.9, 1.1, i_n_examples)
tx = numpy.int32(
numpy.round(numpy.random.uniform(-20, 20, i_n_examples)))
ty = numpy.int32(
numpy.round(numpy.random.uniform(-20, 20, i_n_examples)))
x = self.__roi.x + tx
y = self.__roi.y + ty
#FIXME: Extend valid indices to outliers due to affine transform!!!
min_x = 0
min_y = 0
max_x = i_image.width - self.__roi.width - 1
max_y = i_image.height - self.__roi.height - 1
valid_x = numpy.hstack(
[numpy.nonzero(x >= min_x)[0],
numpy.nonzero(x < max_x)[0]])
valid_y = numpy.hstack(
[numpy.nonzero(y >= min_y)[0],
numpy.nonzero(y < max_y)[0]])
valid_idx = numpy.unique(numpy.hstack([valid_x, valid_y]))
original_roi = cv.cvRect(self.__roi.x, self.__roi.y, self.__roi.width,
self.__roi.height)
if self.__rot_mat == None:
original_rot_matrix = None
else:
original_rot_matrix = cv.cvCloneImage(self.__rot_mat)
for index in valid_idx:
params = numpy.array(
[tx[index], ty[index], scales[index], angles[index]])
self.setAffineTransform(center, scales[index], angles[index])
self.__roi.x = int(x[index])
self.__roi.y = int(y[index])
image = cv.Ipl2NumPy(self.normalise(i_image))
o_data = numpy.dstack([o_data, image])
o_transforms = numpy.vstack([o_transforms, params])
#Restore the original region of interest
self.__roi.x = original_roi.x
self.__roi.y = original_roi.y
if original_rot_matrix == None:
self.__rot_mat = None
else:
self.__rot_mat = cv.cvCloneImage(original_rot_matrix)
return (o_data, o_transforms)
def setRoi(self, i_roi):
self.__roi = cv.cvRect( i_roi.x, i_roi.y, i_roi.width, i_roi.height )
def getRoi(self):
if self.__roi is None:
return None
return cv.cvRect(self.__roi.x, self.__roi.y, self.__roi.width, self.__roi.height)