def loadImage(path, basename='', color='rgb'):
''' Load JPEG image from file,
Returns
--------
IM : 2D or 3D array, size H x W x nColors
dtype will be float64, with each pixel in range (0,1)
'''
path = str(path)
if len(basename) > 0:
path = os.path.join(path, basename)
if color == 'gray' or color == 'grey':
IM = imread(path, as_grey=True)
assert IM.ndim == 2
else:
IM = imread(path, as_grey=False)
if not IM.ndim == 3:
raise ValueError('Color image not available.')
if IM.dtype == np.float:
MaxVal = 1.0
elif IM.dtype == np.uint8:
MaxVal = 255
else:
raise ValueError("Unrecognized dtype: %s" % (IM.dtype))
assert IM.min() >= 0.0
assert IM.max() <= MaxVal
IM = np.asarray(IM, dtype=np.float64)
if MaxVal > 1:
IM /= MaxVal
return IM
def main():
imgs = MultiImage(data_dir + '/multipage.tif')
for a, i in zip(range(0, 4), [1, 9, 7, 8]):
fig = plt.figure()
ax = fig.add_axes([-0.1, -0.1, 1.2, 1.2])
# ax.set_axis_off()
im = data.imread('samolot0' + str(i) + '.jpg', as_grey = True)
im = invert(im)
im = process(im)
out = np.ones_like(im)
io.imshow(out)
contours = measure.find_contours(im, 0.9)
for n, contour in enumerate(contours):
plt.plot(contour[:, 1], contour[:, 0], linewidth=2, color = 'white')
plt.savefig(str(a) + '.jpg', bbox_inches = 0, frameon = False)
fig = plt.figure()
grid = AxesGrid(fig, rect = (1, 1, 1), nrows_ncols = (2, 2), axes_pad = 0.1)
for i in range(0, 4):
frame = data.imread(str(i) + '.jpg')
grid[i].imshow(frame)
grid[i].set_xticks([])
grid[i].set_yticks([])
plt.savefig('na3.jpg')
def load(pos_dir, neg_dir, hard_examples=[]):
gc.disable()
images, labels = [], []
files = listdir(pos_dir)
for i, f in enumerate(files):
if not is_image(f):
continue
if "hard_example" in f and (
hard_examples is None or not any([f.startswith(prefix) for prefix in hard_examples])
):
continue
if i > 0 and i % 1000 == 0:
print "loaded {0}/{1} positive".format(i, len(files))
images.append(data.imread(join(pos_dir, f))[:, :, :3])
labels.append(1)
files = listdir(neg_dir)
for i, f in enumerate(files):
if not is_image(f):
continue
if "hard_example" in f and (
hard_examples is None or not any([f.startswith(prefix) for prefix in hard_examples])
):
continue
if i > 0 and i % 1000 == 0:
print "loaded {0}/{1} negative".format(i, len(files))
images.append(data.imread(join(neg_dir, f))[:, :, :3])
labels.append(0)
gc.enable()
return images, labels
def read_files(dir, dataset):
"""
yield data_type(train? val? test?), numpy.ndarray('uint8')
"""
dir_path = os.path.join(dir,dataset)
if dataset=='train':
for(root, dirs, files) in os.walk(dir_path):
for file in files:
if not '.txt' in file:
label = file.split("_")[0]
img_filepath = os.path.join(root,file)
yield label, data.imread(img_filepath)
elif dataset=='val':
for(root, dirs, files) in os.walk(dir_path):
for file in files:
if '.txt' in file:
# this is val_annotaions.txt
f = open(os.path.join(root,file), 'r')
while 1:
line = f.readline()
if not line: break
line_seg = line.split()
img_filepath = os.path.join(root,'images',line_seg[0])
label = line_seg[1]
yield label, data.imread(img_filepath)
f.close()
def main():
for file_path in glob.glob("/home/lucas/Downloads/Lucas/GSK 10uM/*.JPG"):
img = data.imread(file_path, as_grey=True)
img = transform.resize(img, [600, 600])
img_color = transform.resize(data.imread(file_path), [600, 600])
img[img >img.mean()-0.1] = 0
# io.imshow(img)
# io.show()
#
edges = canny(img)
bordas_fechadas = closing(img > 0.1, square(15)) # fechando gaps
fill_cells = ndi.binary_fill_holes(bordas_fechadas)
# io.imshow(fill_cells)
# io.show()
img_label = label(fill_cells, background=0)
n= 0
for x in regionprops(img_label):
if x.area < 2000 and x.area > 300:
n +=1
print x.area
minr, minc, maxr, maxc = x.bbox
try:
out_path_name = file_path.split("/")[-1].rstrip(".JPG")
io.imsave("out/cell_{}_pic_{}_area_{}.png".format(n, out_path_name, str(round(x.area))),img_color[minr-3: maxr+3, minc-3: maxc+3])
#io.show()
except:
pass
def main():
plt.figure(figsize=(25, 24))
planes = ['samolot00.jpg', 'samolot01.jpg', 'samolot03.jpg', 'samolot04.jpg', 'samolot05.jpg','samolot07.jpg',
'samolot08.jpg', 'samolot09.jpg', 'samolot10.jpg', 'samolot11.jpg', 'samolot12.jpg', 'samolot13.jpg',
'samolot14.jpg', 'samolot15.jpg', 'samolot16.jpg', 'samolot17.jpg', 'samolot18.jpg', 'samolot20.jpg']
i = 1
for file in planes:
img = data.imread(file, as_grey=True)
img2 = data.imread(file)
ax = plt.subplot(6, 3, i)
ax.axis('off')
img **= 0.4
img = filter.canny(img, sigma=3.0)
img = morphology.dilation(img, morphology.disk(4))
img = ndimage.binary_fill_holes(img)
img = morphology.remove_small_objects(img, 1000)
contours = measure.find_contours(img, 0.8)
ax.imshow(img2, aspect='auto')
for n, contour in enumerate(contours):
ax.plot(contour[:, 1], contour[:, 0], linewidth=1.5)
center = (sum(contour[:, 1])/len(contour[:, 1]), sum(contour[:, 0])/len(contour[:, 0]))
ax.scatter(center[0], center[1], color='white')
i += 1
plt.savefig('zad2.pdf')
def Comparer_image_plot(NumImg):
liste1 = os.listdir("images")
for i in liste1:
if str(NumImg) in i:
ImgPolyp = i
break
liste2 = os.listdir("graph_images12")
for j in liste2:
if str(NumImg) in j:
PlotPolyp = j
break
polDia = data.imread("images\\"+ ImgPolyp)
polDia_plot = data.imread("graph_images12\\"+ PlotPolyp)
plt.subplot(1,2,1)
plt.title("Image initiale:" + ImgPolyp)
plt.imshow(polDia)
plt.subplot(1,2,2)
plt.title("Graph image apres analyse:"+ PlotPolyp)
plt.imshow(polDia_plot)
plt.show()
def read_image(name, size=None, debug=False):
""" read image and segmentation, returns RGB + alpha composite """
image = imread(name) / 255.
if image.shape[2] == 4:
alpha = image[...,3]
image = image[...,:3]
else:
segmentation_name = os.path.splitext(name)[0][:-6] + '-label.png'
segmentation = imread(segmentation_name)
alpha = np.logical_or(segmentation[...,0], segmentation[...,1]) * 1.
if size is not None:
scale_x = float(size[0]) / image.shape[1]
scale_y = float(size[1]) / image.shape[0]
scale = min(scale_x, scale_y)
if debug:
print name, size[0], size[1], image.shape[1], image.shape[0], scale, image.shape[1]*scale, image.shape[0]*scale
if scale > 1.0:
print 'Image %s smaller than requested size' % name
if scale != 1.0:
image = rescale(image, scale, order=3)
alpha = rescale(alpha, scale, order=0)
return np.dstack((image, alpha))
def process():
data = {
0: 0,
1: 0,
2: 0,
3: 0,
4: 0,
5: 0,
6: 0,
7: 0,
8: 0,
}
sobel_v_image = imread('sobel_v2.jpg')
sobel_h_image = imread('sobel_h2.jpg')
canny_image = imread('canny2.jpg')
row, col = canny_image.shape
for r in xrange(row):
for c in xrange(col):
if sobel_v_image[r][c] < 0 or sobel_h_image[r][c] < 0:
print sobel_v_image[r][c], sobel_h_image[r][c]
if canny_image[r][c] != 255:
continue
interval = which_interval(sobel_v_image[r][c], sobel_h_image[r][c])
data[interval] += 1
print data
def xformImage(filename, doPlot=False):
#config area
sigma=4
mean_multiplier=2
#end config area
# img_gray = misc.imread(filename, True)
img_gray = imread(filename, as_grey=True)
img_color = imread(filename)
# print(img_color.shape)
# img_gray = resize(skimage.img_as_float(img_gray), (400,400))
img_color_masked = copy.deepcopy(img_color)
# img_color_masked = resize(img_color_masked, (400,400))
# img_color_resized = resize(img_color, (200,200))
# img_color = misc.imread(filename, False)
img_gray = ndimage.gaussian_filter(img_gray, sigma=sigma)
m,n = img_gray.shape
# sx = ndimage.sobel(img_gray, axis=0, mode='constant')
# sy = ndimage.sobel(img_gray, axis=1, mode='constant')
# sob = np.hypot(sx, sy)
mask = (img_gray > img_gray.mean()*mean_multiplier).astype(np.float)
labels = morphology.label(mask)
center_label = labels[m/2, m/2]
labels[labels != center_label] = 0
labels[labels == center_label] = 1
# img_test = copy.deepcopy(img_color)
# img_test[ labels == 0, : ] = 0
# sob [ labels == 0] = 0
img_color_masked [labels == 0, :] = 0
# img_test = ndimage.gaussian_filter(img_test, 3)
if doPlot:
f, (ax_gray, ax_color, ax_sob) = plt.subplots(ncols=3)
ax_gray.imshow(img_color_masked, cmap=plt.cm.get_cmap('gray'))
ax_gray.axis('off')
# ax_sob.imshow(sob, cmap=plt.cm.get_cmap('gray'))
ax_sob.axis('off')
ax_color.imshow(img_color, cmap=plt.cm.get_cmap('gray'))
ax_color.axis('off')
plt.show()
return np.reshape(img_color_masked, -1)
def load_image(path, as_grey=False, to_float=True):
if DEBUG:
im = imread(path, as_grey)
im = (im - np.amin(im) * 1.0) / (np.amax(im) - np.amin(im))
return im
# Load image
image = imread(path, as_grey)
if to_float:
# Convert to floating point matrix
image = image.astype(np.float32)
return image
def pestFeatureExtraction(filename):
selem = disk(8)
image = data.imread(filename,as_grey=True)
thresh = threshold_otsu(image)
elevation_map = sobel(image)
markers = np.zeros_like(image)
if ((image<thresh).sum() > (image>thresh).sum()):
markers[image < thresh] = 1
markers[image > thresh] = 2
else:
markers[image < thresh] = 2
markers[image > thresh] = 1
segmentation = morphology.watershed(elevation_map, markers)
segmentation = dilation(segmentation-1, selem)
segmentation = ndimage.binary_fill_holes(segmentation)
segmentation = np.logical_not(segmentation)
image[segmentation]=0;
hist = np.histogram(image.ravel(),256,[0,1])
hist = list(hist[0])
hist[:] = [float(x) / (sum(hist) - hist[0]) for x in hist]
hist.pop(0)
features = np.empty( (1, len(hist)), 'float' )
a = np.array(list(hist))
f = a.astype('float')
features[0,:]=f[:]
return features
def test_iterations(n_steps, l, n_detectors, filename, labelname, filtr):
image = data.imread("os.png", as_grey=True)
image /= np.max(image)
per_steps = np.arange(0, 361, 20)
per_steps[len(per_steps) - 1] = 360
result = np.zeros(len(per_steps))
for i, j in enumerate(per_steps):
if j == 0:
j += 1
sin = createSinogram(image, j, l, n_detectors)
sin2 = sin.copy()
sin2 /= np.max(sin2)
if filtr == True:
sin2 = sinogram_filter(sin2)
invImage = createImage(sin2, j, l, n_detectors, 200)
else:
invImage = createImage(sin2, j, l, n_detectors, 200)
cpy = invImage.copy()
if filtr == True:
for x in range(len(cpy)):
for y in range(len(cpy[x])):
if cpy[x][y] != 0:
cpy[x][y] = abs(cpy[x][y])
cpy /= np.max(cpy)
result[i] = ret_dif_bet_2_inputs(image, cpy)
savetest(per_steps, result, filename, labelname)
def loadFile(self):
self.path = filedialog.askopenfilename()
if self.path == () or self.path == "":
self.path = ""
return
self.LoadLabel.config(text=self.path)
fileExtension = self.path.split(".")
if len(fileExtension) < 2 or not fileExtension[1] == "dcm":
newImage = data.imread(self.path, as_grey=True)
self.images.changePlot((2, 3, 1), newImage)
return
lastNamez, firstNamez, id, birthdayz, sexz, datez, timez, newImage = readDicomData(
self.path)
self.images.changePlot((2, 3, 1), newImage)
self.lastNameDicom.delete("1.0", tk.END)
self.lastNameDicom.insert("1.0", lastNamez)
self.NameDicom.delete("1.0", tk.END)
self.NameDicom.insert("1.0", firstNamez)
self.idDicom.delete("1.0", tk.END)
self.idDicom.insert("1.0", id)
self.dateDicom.config(text=datez)
self.timeDicom.config(text=timez)
self.birthh.delete("1.0", tk.END)
self.birthh.insert("1.0", birthdayz)
sex = sexz.upper()
if sex == "M":
self.sexx.current(0)
elif sex == "F":
self.sexx.current(1)
def main():
inData = data.imread("resources/in.png", as_grey=True)
plt.subplot(2, 3, 1)
plt.title("Original image")
plt.imshow(inData, cmap='gray')
plt.subplot(2, 3, 2)
plt.xlabel("Emiter/detector rotation")
plt.ylabel("Number of receiver")
plt.title("Radon transform image")
radonImage = radonTransform(inData,
stepSize=step,
detectorsNumber=detectorsNumber,
detectorsWidth=detectorWidth)
plt.imshow(radonImage,
cmap='gray',
extent=[0, 180, len(radonImage), 0],
interpolation=None)
inverseRadonImage = inverseRadonTransform(radonImage,
stepSize=step,
detectorsWidth=detectorWidth)
saveDicomFile("out.dcm", "Andrzej", "1234", "male", "20070304",
inverseRadonImage)
#write_dicom(inverseRadonImage,'pretty.dcm')
plt.subplot(2, 3, 4)
plt.title("Inverse Radon transform image")
plt.imshow(inverseRadonImage, cmap='gray')
plt.show()
def segment_slic(path, n_segments=N_SEGMENTS):
img = img_as_float(imread(path))
segments = slic(img, n_segments=n_segments, compactness=10, sigma=1)
# segments = watershed(img, markers=250, compactness=0.1)
# segments = quickshift(img, kernel_size=3, max_dist=6, ratio=0.5)
# segments = felzenszwalb(img, scale=100, sigma=0.5, min_size=50)
return img, segments
def main(argv):
# lendo imagem
image1 = img_as_float(data.imread("data/" + argv[0]))
channels = 'gray'
# verificando número de canais
if len(image1.shape) == 3:
channels = None
# imagem borrada com filtro gaussiano
img_borrada = filters.gaussian_filter(image1, sigma=3.0)
# diferença da imagem original com a borrada
g_mascara = np.subtract(image1, img_borrada)
# high boost
g_image = np.add(np.multiply(2, g_mascara), image1)
# salvando imagem
plt.imsave(argv[1], g_image, cmap=channels) #saving
# plotagem
plt.figure().suptitle("filtro da mediana")
plt.subplot(1, 2, 2)
plt.imshow(g_image, cmap=channels)
plt.title('Depois')
plt.axis('off')
# Divide a área de plotagem: 1 linha e 3 colunas.
plt.subplot(1, 2, 1) # A área ativa é a 1.
plt.imshow(image1, cmap=channels)
plt.title('Antes')
plt.axis('off')
plt.show()
def __init__(self, im_path):
"""Accepts the path to a tif or tiff file,
then attempts to output a numpy uint8 array
that has the shape (z, y, x, color)"""
if not path.isfile(im_path):
raise IOError("File does not exist")
if not im_path.lower().endswith(('.tif', '.tiff')):
raise IOError("File type incorrect")
self.path = im_path
self.im = data.imread(im_path)
self.im_shape = self.im.shape
self.new_im = self.im
color_axis = np.argmin(self.im_shape)
if color_axis == 1:
self.new_im = []
print("Restacking array... current size: {}".format(str(self.im_shape)))
for z_slice in range(self.im_shape[0]):
for channel in range(self.im_shape[color_axis]):
s_data = self.im[z_slice]
self.new_im.append(np.dstack((s_data[0], s_data[1], s_data[2])))
self.new_im = np.array(self.new_im, dtype="uint8")
print("Restack completed. current size: {}".format(str(self.new_im.shape)))
if color_axis != 1 and color_axis != 3:
raise ValueError("File format not recognized")
def getseg(gt_prefix, gt_ext, imname, fg_cat):
path_name, ext = os.path.splitext(imname)
gt_name = os.path.join(gt_prefix, '%s_gt.%s'%(path_name,gt_ext))
#pdb.set_trace()
seg = imread(gt_name)
seg = [seg == fg_cat]
return seg
def images_loader_mini(input_path, scale):
hr_images_train = []
lr_images_train = []
hr_images_test = []
lr_images_test = []
count = 0
file_names = \
[input_path + f for f in listdir(input_path) if (isfile(join(input_path, f)) and not f.startswith('.'))]
print('Loading images')
for file in tqdm(file_names):
image = data.imread(file)
image = set_image_alignment(image, scale)
image_batch = get_split_images(image)
image = resize_image(image, 1 / scale)
image = set_image_alignment(image, scale)
input_image = resize_image(image, 1 / scale)
hr_images_test.append(image)
lr_images_test.append(input_image)
for i in range(0, image_batch.shape[0]):
for j in range(0, image_batch.shape[1]):
input_image = resize_image(image_batch[i][j], 1 / scale)
hr_images_train.append(image_batch[i][j])
lr_images_train.append(input_image)
count += 1
print(str(count) + ' images loaded')
return hr_images_test, lr_images_test, hr_images_train, lr_images_train
def main(argv):
# lendo imagem
image1 = img_as_float(data.imread("data/" + argv[0]))
channels = 'gray'
# verificando número de canais
if len(image1.shape) == 3:
channels = None
# alargamento de contraste
final = exposure.rescale_intensity(image1)
# salvando imagem
plt.imsave(argv[1], final, cmap=channels)
# plotagem
plt.figure().suptitle("Transformação negativa")
plt.subplot(1,2,2)
plt.imshow(final, cmap=channels)
plt.title('Depois')
plt.axis('off')
# Divide a área de plotagem: 1 linha e 3 colunas.
plt.subplot(1,2,1) # A área ativa é a 1.
plt.imshow(image1, cmap=channels)
plt.title('Antes')
plt.axis('off')
plt.show()
def load_data(data_directory):
## Read all subdirectories in the data directory
directories = [
d for d in os.listdir(data_directory)
if os.path.isdir(os.path.join(data_directory, d))
]
## Initialize image and label arrays
images = []
labels = []
## For every subdirectory in the data directory,
## read all jpg images name and append it to the images array,
## and append subdirecoty name to the labels array.
for d in directories:
plant_directory = os.path.join(data_directory, d)
file_names = [
os.path.join(plant_directory, filename)
for filename in os.listdir(plant_directory)
if filename.endswith(".jpg")
]
for filename in file_names:
images.append(data.imread(filename))
labels.append(int(d))
return images, labels
def WeinerFilter():
e1 = cv2.getTickCount()
astro = color.rgb2gray(data.imread(imagePath))
psf = np.ones((5, 5)) / 25
astro = conv2(astro, psf, 'same')
astro += 0.1 * astro.std() * np.random.standard_normal(astro.shape)
deconvolved, _ = restoration.unsupervised_wiener(astro, psf)
fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(8, 5), sharex=True, sharey=True, subplot_kw={'adjustable':'box-forced'})
plt.gray()
ax[0].imshow(astro, vmin=deconvolved.min(), vmax=deconvolved.max())
ax[0].axis('off')
ax[0].set_title('Original Picture')
ax[1].imshow(deconvolved)
ax[1].axis('off')
ax[1].set_title('Self tuned restoration')
fig.subplots_adjust(wspace=0.02, hspace=0.2,
top=0.9, bottom=0.05, left=0, right=1)
e2 = cv2.getTickCount()
time = (e2 - e1)/ cv2.getTickFrequency() + (numberThreads) + numberThreadsPerCore + numberCores
msgbox(msg= str(time) +" seconds", title="Execution Time", ok_button="OK")
plt.show()
def parse(train_gt, train_img_dir, info=False, fast_train=False):
from skimage.data import imread
from scipy.ndimage import zoom
if fast_train:
train_X = np.zeros((500, im_size, im_size, 1))
train_Y = np.zeros((500, coords_size))
else:
train_X = np.zeros((len(train_gt), im_size, im_size, 1))
train_Y = np.zeros((len(train_gt), coords_size))
for i, img_name in enumerate(train_gt):
if i == 500 and fast_train:
break
img = imread(train_img_dir+'/'+img_name, as_grey=True)
train_Y[i] = train_gt[img_name]
for j in range(1, coords_size, 2):
train_Y[i][j] *= im_size/img.shape[0]
for j in range(0, coords_size, 2):
train_Y[i][j] *= im_size/img.shape[1]
img = zoom(img, [im_size/img.shape[0], im_size/img.shape[1]])
img = (img / 255)
train_X[i,:,:,0] = img
del(img)
if info and (i+1)%100 == 0:
print('Image: ', i+1, end='\r')
return train_X, train_Y
def detect(model, test_img_dir):
from os import listdir
from skimage.data import imread
from scipy.ndimage import zoom
img_list = listdir(test_img_dir)
data = np.zeros((100, im_size, im_size, 1))
sizes = []
k = 0
ans = {}
for i, img_name in enumerate(img_list):
img = imread(test_img_dir+'/'+img_name, as_grey=True)
sizes.append([img_name, img.shape])
img = zoom(img, [im_size/img.shape[0], im_size/img.shape[1]])
img = (img / 255)
data[k,:,:,0] = img
k += 1
del(img)
if (i+1)%100 == 0:
print((i+1), ' images')
points = model.predict(data, verbose=1)
for i in range(len(points)):
points[i][1::2] *= sizes[i][1][0]/im_size
points[i][::2] *= sizes[i][1][1]/im_size
ans[sizes[i][0]] = points[i]
sizes = []
k = 0
data = np.zeros((100, im_size, im_size, 1))
if k != 0:
data = data[:k,:,:,:]
points = model.predict(data, verbose=1)
for i in range(len(points)):
points[i][1::2] *= sizes[i][1][0]/im_size
points[i][::2] *= sizes[i][1][1]/im_size
ans[sizes[i][0]] = points[i]
return ans
def process(q, iolock):
global product_count
global category_count
global picture_count
while True:
d = q.get()
if d is None:
break
product_count += 1
product_id = str(d['_id'])
category_id = str(d['category_id'])
category_dir = os.path.join(images_dir, category_id)
if not os.path.exists(category_dir):
category_count += 1
try:
os.makedirs(category_dir)
except:
pass
for e, pic in enumerate(d['imgs']):
picture_count += 1
picture = imread(io.BytesIO(pic['picture']))
picture_file = os.path.join(category_dir,
product_id + '_' + str(e) + '.jpg')
if not os.path.isfile(picture_file):
plt.imsave(picture_file, picture)
def main(argv):
# lendo imagem
image1 = data.imread("data/" + argv[0])
# plotando histograma
plt.hist((image1).flatten(), density=True, color='purple', bins=25)
plt.show()
def blob_counter(img, min_sigma=2, max_sigma=8, treshold=0.0001, overlap=0.6):
img = imread(img, as_grey=True)
bw = img.mean(axis=2)
blobs_dog = [(x[0], x[1], x[2]) for x in feature.blob_dog(
-bw, min_sigma=2, max_sigma=8, threshold=0.0001, overlap=0.6)]
blobs_dog = set(blobs_dog)
return str(len(blobs_dog))
def load_scenes(filename):
zipped_scenes = []
print 'Working on: ' + filename
img = data.imread('scenes/' + filename, as_grey=True)
tmp = img
tmp = filter.canny(tmp, sigma=2.0)
tmp = ndimage.binary_fill_holes(tmp)
#tmp = morphology.dilation(tmp, morphology.disk(2))
tmp = morphology.remove_small_objects(tmp, 2000)
contours = measure.find_contours(tmp, 0.8)
ymin, xmin = contours[0].min(axis=0)
ymax, xmax = contours[0].max(axis=0)
if xmax - xmin > ymax - ymin:
xdest = 1000
ydest = 670
else:
xdest = 670
ydest = 1000
src = np.array(((0, 0), (0, ydest), (xdest, ydest), (xdest, 0)))
dst = np.array(((xmin, ymin), (xmin, ymax), (xmax, ymax), (xmax, ymin)))
tform3 = tf.ProjectiveTransform()
tform3.estimate(src, dst)
warped = tf.warp(img, tform3, output_shape=(ydest, xdest))
tmp = filter.canny(warped, sigma=2.0)
tmp = morphology.dilation(tmp, morphology.disk(2))
descriptor_extractor.detect_and_extract(tmp)
obj_key = descriptor_extractor.keypoints
scen_desc = descriptor_extractor.descriptors
zipped_scenes.append([warped, scen_desc, obj_key, filename])
return zipped_scenes
def read_image(image_path, image_shape=None, preserve_range=True):
image = imread(image_path)
if image_shape:
image = resize(image, image_shape, preserve_range=preserve_range)
image = image[None]
return image
def show_train_data():
"""
Plots original image, mask and image + mask
"""
masks = train
for ImageId in os.listdir('train_positives/dev/class1'):
img = imread('train_positives/dev/class1/' + ImageId)
img_masks = masks.loc[masks['ImageId'] == ImageId,
'EncodedPixels'].tolist()
# Take the individual ship masks and create a single mask array for all ships
if type(img_masks[0]) == str:
all_masks = np.zeros((768, 768))
for mask in img_masks:
all_masks += rle_decode(mask)
fig, axarr = plt.subplots(1, 3, figsize=(15, 40))
axarr[0].axis('off')
axarr[1].axis('off')
axarr[2].axis('off')
axarr[0].imshow(img)
axarr[1].imshow(all_masks)
axarr[2].imshow(img)
axarr[2].imshow(all_masks, alpha=0.4)
plt.tight_layout(h_pad=0.1, w_pad=0.1)
#savefig(ImageId)
plt.show()
else:
print('NaN')
def main():
def shutdown(signal, widget):
exit()
signal.signal(signal.SIGINT, shutdown)
app = QtGui.QApplication(sys.argv)
timer = QtCore.QTimer()
timer.start(500)
timer.timeout.connect(lambda: None)
loaded = lh.load_files(str(DATA_PATH), filter_unfinished=False)
for l in loaded:
jpg_io = io.BytesIO(l['jpg_image'])
l['image'] = imread(jpg_io)
l['save_path'] = os.path.join(DATA_PATH, l['filename'])
if 'edited' not in l:
l['edited'] = False
dh.triangulation(
loaded,
do_triangulation=tr.triangulation,
do_import=partial(dh.import_bone, SEGMENTATION_METHODS)
)
sys.exit(app.exec_())
def show_result(eye, manual_segmentation):
plt.imshow(data.imread(sys.argv[1]), cmap=plt.gray())
plt.show()
plt.imshow(eye, cmap=plt.gray())
plt.show()
plt.imshow(manual_segmentation, cmap=plt.gray())
plt.show()
def ucf101():
data_path = '/home/sl/Resource/UCF101/'
store_path = './list/UCF101/'
random_num = 4
# make train.txt and val.txt
data_dirs = sorted(os.listdir(data_path))
train_list = []
val_list = []
label = -1
count_for_train = 0
count_for_val = 0
mean_list = []
label_list = []
for data_dir in data_dirs:
video_path = os.path.join(data_path, data_dir)
video_dirs = sorted(os.listdir(video_path))
label += 1
label_list.append(data_dir)
for video_dir in video_dirs:
print("now for data dir %s\n" % video_dir)
img_path = os.path.join(video_path, video_dir)
if not random.randint(0, random_num) == random_num:
train_list.append((img_path, label))
count_for_train += 1
else:
val_list.append((img_path, label))
count_for_val += 1
# calculate mean of img
imgs = sorted(os.listdir(img_path))
for img in imgs:
pic_path = os.path.join(img_path, img)
pic = ski.imread(pic_path)
mean_list.append(np.mean(pic, (0, 1)))
os.chdir(store_path)
train_file = open('train.txt', 'w')
for item in train_list:
train_file.write("%s %s \n" % (item[0], item[1]))
train_file.close()
val_file = open('val.txt', 'w')
for item in val_list:
val_file.write("%s %s \n" % (item[0], item[1]))
val_file.close()
label_file = open('label.txt', 'w')
for item in label_list:
label_file.write("%s \n" % item)
label_file.close()
mean = np.mean(np.array(mean_list), 0)
others_file = open('others.txt', 'w')
others_file.write("mean : %.2f %.2f %.2f\n" % (mean[0], mean[1], mean[2]))
others_file.write("train num: %d\n" % count_for_train)
others_file.write("val num: %d\n" % count_for_val)
print("train num: ", count_for_train)
print("val num: ", count_for_val)
print("mean val: ", mean)
def find_alpha(base_img, img, model_robust):
# what type of interpolation
# 0: nearest-neighbor
# 1: bi-linear
warp_order = 1
output_shape, corner_min = find_output_shape(base_img, model_robust, channel)
#print("output_shape", output_shape, corner_min)
#print(model_robust.scale, model_robust.translation, model_robust.rotation)
# This in-plane offset is the only necessary transformation for the base image
offset = SimilarityTransform(translation= -corner_min)
base_warped = warp(base_img[:,:,channel], offset.inverse, order=warp_order,
output_shape = output_shape, cval=-1)
base_color = warp(base_img, offset.inverse, order=warp_order,
output_shape = output_shape, cval=-1)
# warp image corners to new position in mosaic
transform = (model_robust + offset).inverse
#img_warped = warp(img[:,:,channel], transform, order=warp_order,
# output_shape=output_shape, cval=-1)
img_color = warp(img, transform, order=warp_order,
output_shape=output_shape, cval=-1)
#base_mask = (base_warped != -1)
#base_warped[~base_mask] = 0
img_mask = (img_warped != -1)
#img_warped[~img_mask] = 0
#convert to rgb
base_alpha = add_alpha(base_color, base_mask)
img_alpha = np.dstack((img_color, img_mask))
#base_alpha = np.dstack((base_color, base_mask))
#plt.imsave(tmp_base, base_alpha )
#plt.imsave(tmp_img, img_alpha )
#cmd = [path_to_enblend, tmp_base, tmp_img, '-o', tmp_out]
#p = Popen(cmd, stdin=PIPE, stdout=PIPE, stderr=PIPE)
#output, err = p.communicate(b"input data that is passed to subprocess' stdin")
#rc = p.returncode
# remove alpha channel
if os.path.exists(tmp_out):
out = imread(tmp_out)[:,:,:3]
else:
print("couldnt find out image")
print(rc, output, err)
plt.figure()
plt.imshow(base_alpha)
plt.figure()#
plt.imshow(img_alpha)
plt.show()
out = base_alpha[:,:,:3]
#if you don't have enblend, you can use one of these
#merged_img = simple_merge(base_warped, img_warped, base_mask, img_mask)
#merged_img = minimum_cost_merge(base_warped, img_warped, base_mask, img_mask)
#merged_edges = remove_empty_edges(merged_img)
return tmp_alpha
def main(argv):
# definindo mascara laplaciana com centro -4
lap_4_ = np.array([[0., 1., 0.], [1., -4., 1.], [0., 1., 0.]], dtype=float)
image1 = img_as_float(data.imread("data/" + argv[0]))
channels = 'gray'
# verificando numero de canais
if len(image1.shape) == 3:
channels = None
lap_4_ = lap_4_[:, :, None]
# aplicando convolução
final = filters.convolve(image1, lap_4_)
# correção de intensidades
final_corr = final + image1
# salvando imagem
plt.imsave(argv[1], final, cmap=channels)
# plotando resultados
plt.figure().suptitle("Aguçamento laplaciano")
plt.subplot(1, 2, 2)
plt.imshow(final_corr, cmap=channels)
plt.title('Depois')
plt.axis('off')
plt.subplot(1, 2, 1)
plt.imshow(final, cmap=channels)
plt.title('Antes')
plt.axis('off')
plt.show()
def load_data(data_directory):
print("Inside Load_Data")
directories = [
d for d in os.listdir(data_directory)
if os.path.isdir(os.path.join(data_directory, d))
]
print("directories\n:", directories)
labels = []
images = []
pdb.set_trace()
for d in directories:
label_directory = os.path.join(data_directory, d)
#print("label_directory:", label_directory)
file_names = [
os.path.join(label_directory, f)
for f in os.listdir(label_directory) if f.endswith(".ppm")
]
#pdb.set_trace()
for f in file_names:
#images.append(skimage.data.imread(f))
images.append(data.imread(f))
labels.append(int(d))
#pdb.set_trace()
return images, labels
def just_do_it(limit_cont):
fig = plt.figure(facecolor='black')
plt.gray()
print("Rozpoczynam przetwarzanie obrazow...")
for i in range(20):
img = data.imread(images[i])
gray_img = to_gray(images[i]) # samoloty1.pdf
#gray_img = to_gray2(images[i], 1001, 0.2, 5, 9, 12) # samoloty2.pdf
#gray_img = to_gray2(images[i], 641, 0.2, 5, 20, 5) # samoloty3.pdf
conts = find_contours(gray_img, limit_cont)
centrs = [find_centroid(cont) for cont in conts]
ax = fig.add_subplot(4,5,i)
ax.set_yticks([])
ax.set_xticks([])
io.imshow(img)
print("Przetworzono: " + images[i])
for n, cont in enumerate(conts):
ax.plot(cont[:, 1], cont[:, 0], linewidth=2)
for centr in centrs:
ax.add_artist(plt.Circle(centr, 5, color='white'))
fig.tight_layout()
#plt.show()
plt.savefig('samoloty3.pdf')
def process(filename):
img = data.imread(filename + '.png')
f = open(filename + '.txt', 'w')
for x in range(len(img)):
for y in range(len(img[x])):
if img[x][y] == 0:
f.write(str(y) + ',' + str(x) + '\n')
def make_prediction():
import sys
import numpy as np
import pandas as pd
from skimage.data import imread
from skimage.filters import threshold_adaptive
from skimage.restoration import denoise_tv_bregman
from sklearn.cross_validation import train_test_split
from sklearn.grid_search import GridSearchCV
from sklearn.neighbors import KNeighborsClassifier
from sklearn.ensemble import RandomForestClassifier, AdaBoostClassifier
from model_design import model_design
classifier = model_design()
X, IDs = [], range(6284, 12504)
for ID in IDs:
original = imread('../data/testResized/' + str(ID) +'.Bmp', as_grey=True)
denoised = denoise_tv_bregman(original, 3)
binarilized = threshold_adaptive(denoised, block_size=13, method='gaussian')
feature = binarilized.reshape(1,400)[0]
X.append(feature)
X = np.array(X)
y = classifier.predict(X)
result = pd.DataFrame({'Id': IDs, 'Class': y})
result.to_csv('../result/06-09-2015_AdaBoostXTC.csv', sep=',', index=None, columns=['Id', 'Class'])
def main(argv):
# lendo imagem
image1 = img_as_float(data.imread("data/" + argv[0]))
channels = 'gray'
# verificando número de canais
if len(image1.shape) == 3:
channels = None
# aplicando filtro gaussiano
final = filters.gaussian_filter(image1, sigma=float(argv[2]))
# salvando imagem
plt.imsave(argv[1], final, cmap=channels)
# plotagem
plt.figure().suptitle("filtro gaussiano")
plt.subplot(1, 2, 2)
plt.imshow(final, cmap=channels)
plt.title('Depois')
plt.axis('off')
# Divide a área de plotagem: 1 linha e 3 colunas.
plt.subplot(1, 2, 1) # A área ativa é a 1.
plt.imshow(image1, cmap=channels)
plt.title('Antes')
plt.axis('off')
plt.show()
def main(argv):
# lendo imagem
image1 = img_as_float(data.imread("data/" + argv[0]))
channels = 'gray'
# verificando número de canais
if len(image1.shape) == 3:
channels = None
# transformação de potência
final = np.power(image1, float(argv[2]))
# salvando imagem
plt.imsave(argv[1], final, cmap=channels)
# plotagem
plt.figure().suptitle("Transformação gama")
plt.subplot(1,2,2)
plt.imshow(final, cmap=channels)
plt.title('Depois')
plt.axis('off')
# Divide a área de plotagem: 1 linha e 3 colunas.
plt.subplot(1,2,1) # A área ativa é a 1.
plt.imshow(image1, cmap=channels)
plt.title('Antes')
plt.axis('off')
plt.show()
def chooseFile(self):
# self.images.config(state="readonly")
self.path = filedialog.askopenfilename()
if self.path == () or self.path == "":
self.path = ""
return
self.pathLabel.config(text=self.path)
# print(self.path)
fileExtension = self.path.split(".")
if len(fileExtension) < 2 or not fileExtension[1] == "dcm":
# self.images["values"] = fileExtension[0].split("/")[-1]
newImage = data.imread(self.path, as_grey=True)
self.graph.changePlot((2, 2, 1), newImage)
return
lastName, firstName, id, birthday, gender, date, time, newImage = file.readDicomFileToNumpyArray(
self.path)
self.graph.changePlot((2, 2, 1), newImage)
self.lastNameDicom.delete("1.0", tk.END)
self.lastNameDicom.insert("1.0", lastName)
self.firstNameDicom.delete("1.0", tk.END)
self.firstNameDicom.insert("1.0", firstName)
self.idDicom.delete("1.0", tk.END)
self.idDicom.insert("1.0", id)
self.dateDicom.config(text=date)
self.timeDicom.config(text=time)
self.birthDicom.delete("1.0", tk.END)
self.birthDicom.insert("1.0", birthday)
gender = gender.upper()
if gender == "M":
self.genderDicom.current(0)
elif gender == "F":
self.genderDicom.current(1)
def test_save_buttons():
viewer = get_image_viewer()
sv = SaveButtons()
viewer.plugins[0] += sv
import tempfile
fid, filename = tempfile.mkstemp(suffix='.png')
os.close(fid)
timer = QtCore.QTimer()
timer.singleShot(100, QtGui.QApplication.quit)
# exercise the button clicks
sv.save_stack.click()
sv.save_file.click()
# call the save functions directly
sv.save_to_stack()
with expected_warnings(['precision loss']):
sv.save_to_file(filename)
img = data.imread(filename)
with expected_warnings(['precision loss']):
assert_almost_equal(img, img_as_uint(viewer.image))
img = io.pop()
assert_almost_equal(img, viewer.image)
os.remove(filename)
def load_data(data_dir):
sub_dirs = [
d for d in os.listdir(data_dir)
if os.path.isdir(os.path.join(data_dir, d))
]
# labels and images
labels = []
images = []
# extract image files and labels
for d in sub_dirs:
label_dir = os.path.join(data_dir, d) # extract directory
filenames = [
os.path.join(label_dir, f) for f in os.listdir(label_dir)
if f.endswith(".ppm")
]
# create list of file names and labels
for f in filenames:
images.append(data.imread(f))
labels.append(int(d))
# return images with their labels
return labels, images
def make_images(input_file, output_dir):
data = bson.decode_file_iter(open(input_file, 'rb'))
with warnings.catch_warnings():
for item in tqdm(data):
product_id = item['_id']
for ix, pic in enumerate(item['imgs']):
warnings.filterwarnings('error')
try:
img = imread(io.BytesIO(pic['picture']))
imsave(f"{output_dir}/{product_id}_{ix}.jpg", img)
except Warning:
warnings.filterwarnings('ignore')
img = imread(io.BytesIO(pic['picture']))
imsave(f'{output_dir}/low_contrast/{product_id}_{ix}.jpg',
img)
imsave(f"{output_dir}/{product_id}_{ix}.jpg", img)
def load_data(path):
folder = []
images = []
labels = []
for a, b, c in os.walk(ROOT_PATH):
if a.split('/')[-1][0:5] == 'train':
folder.append(a)
for i in range(len(folder)):
#print(folder[i])
one_img = []
one_csv = []
for a1, b1, c1 in os.walk(folder[i]):
for f in c1:
if f.split('.')[-1] == 'jpg':
one_img.append(os.path.join(a1, f))
elif f.split('.')[-1] == 'csv':
one_csv.append(os.path.join(a1, f))
for j in range(len(one_img)):
images.append(data.imread(one_img[j]))
csv_file = one_csv[0]
ooo = pd.read_csv(csv_file, sep=';')
#print(ooo.shape[0])
for i in range(ooo.shape[0]):
labels.append(ooo['Label'][i])
print('Load Suecced')
return images, labels
def test_save_buttons():
viewer = get_image_viewer()
sv = SaveButtons()
viewer.plugins[0] += sv
import tempfile
fid, filename = tempfile.mkstemp(suffix='.png')
os.close(fid)
timer = QtCore.QTimer()
timer.singleShot(100, QtGui.QApplication.quit)
# exercise the button clicks
sv.save_stack.click()
sv.save_file.click()
# call the save functions directly
sv.save_to_stack()
with expected_warnings(['precision loss']):
sv.save_to_file(filename)
img = data.imread(filename)
with expected_warnings(['precision loss']):
assert_almost_equal(img, img_as_uint(viewer.image))
img = io.pop()
assert_almost_equal(img, viewer.image)
os.remove(filename)
def testGenerator():
# test all images in the directory
assert os.path.exists(test_dir), "local image path doesnt exist"
imgs = []
for p in getPaths(test_dir):
# read and scale inputs
img = data.imread(p, as_grey=False)
img = trans.resize(img, im_shape)
img = np.expand_dims(img, axis=0)
# inference
out_img = model.predict(img)
# thresholding
out_img[out_img>0.5] = 1.
out_img[out_img<=0.5] = 0.
print ("tested: {0}".format(p))
# get filename
img_name = ntpath.basename(p)
img_name = img_name.split('.')[0]
# save individual output masks
ROs = np.reshape(out_img[0,:,:,0], (im_h, im_w))
FVs = np.reshape(out_img[0,:,:,1], (im_h, im_w))
HDs = np.reshape(out_img[0,:,:,2], (im_h, im_w))
RIs = np.reshape(out_img[0,:,:,3], (im_h, im_w))
WRs = np.reshape(out_img[0,:,:,4], (im_h, im_w))
misc.imsave(RO_dir+img_name+'.bmp', ROs)
misc.imsave(FB_dir+img_name+'.bmp', FVs)
misc.imsave(HD_dir+img_name+'.bmp', HDs)
misc.imsave(RI_dir+img_name+'.bmp', RIs)
misc.imsave(WR_dir+img_name+'.bmp', WRs)
# combine the masks in a single RGB and save
mask_rgb = get_rgb_from_masks(HDs, ROs, WRs, RIs, FVs)
misc.imsave(samples_dir+img_name+'.bmp', mask_rgb)
def main(argv):
# lendo imagem
image1 = img_as_float(rgb2gray(data.imread("data/" + argv[0])))
channels = 'gray'
# verificando número de canais
if len(image1.shape) == 3:
channels = None
# mascara da média
mascara = np.ones([int(argv[2]), int(argv[2])], dtype='float')
mascara_final = np.divide(mascara, int(argv[2])**2)
# processo de suavização
final = fil.convolve(image1, mascara_final, mode='constant', cval=0)
# aplicando otsu
otsu = filters.threshold_otsu(image1)
final_image = final < otsu
# salvando imagem
plt.imsave(argv[1], final_image, cmap=channels)
# plotagem
plt.figure().suptitle("Suavização + otsu")
plt.subplot(1, 2, 2)
plt.imshow(final_image, cmap=channels)
plt.title('Depois')
plt.axis('off')
# Divide a área de plotagem: 1 linha e 3 colunas.
plt.subplot(1, 2, 1) # A área ativa é a 1.
plt.imshow(image1, cmap=channels)
plt.title('Antes')
plt.axis('off')
plt.show()
def transform_pcf(output_dir="",
patch_size=32,
rotations=4,
limit=None):
training = pd.read_csv("training.csv",
header=None,
names=['name', 'fga'],
dtype={'name': object, 'fga': float})
if limit:
training = training.head(limit)
from skimage.io import imsave
for i, row in training.iterrows():
row_patches = []
row_output = []
name = row['name']
fga = row['fga']
print("Transforming image %s" % name)
import sys
sys.stdout.flush()
for kind in ["DX", "TS"]:
img = imread("images/%s/%s-%s.png" % (kind, name, kind))
img_patches = transform_img(img, name, patch_size=patch_size)
for img_patch in img_patches:
for rot_i in range(rotations):
row_patches.append(np.rot90(img_patch, k=rot_i))
row_output.append(fga)
X_row = np.asarray(row_patches)
y_row = np.asarray(row_output)
with open(path.join(output_dir, "X_file_%s" % name), 'w') as f:
np.save(f, X_row)
with open(path.join(output_dir, "y_file_%s" % name), 'w') as f:
np.save(f, y_row)
def hash(self, files):
for path in files:
try:
p = imread(path)
except IOError as e:
print "oups", e
continue
yield path, dhash(p)
def read_stack(path):
"""Load and return stack as normalized unsigned 8bit ints"""
assert psutil.virtual_memory()[1] > 1000*1024*1024, \
'Less than 1GiB of memory available'
stack = data.imread(path)
stack -= stack.min()
np.true_divide(stack, stack.max() / 255, out=stack, casting='unsafe')
return stack.astype(np.uint8)
def check_train_dim():
rel_path = '../data/images_training_rev1/'
files = os.listdir(rel_path)
for i, f in enumerate(files):
n_s = imread(rel_path + f).shape
if i > 0:
assert(n_s == s)
s = n_s
def read_stack(path):
"""Load and return stack as normalized unsigned 8bit ints"""
assert psutil.phymem_usage()[1] > 1000*1024*1024, \
'Less than 1GiB of memory available'
stack = data.imread(path)
stack -= stack.min()
stack /= stack.max()/255
return stack.astype(np.uint8)
def getContourImage(imageFile):
planeImage = data.imread(imageFile,True)
imageArray = np.asarray(planeImage)
averageColor = np.mean(imageArray)
imageArray = getBlackAndWhiteImage(imageArray,averageColor*0.85)
imageArray = filters.sobel(imageArray)
imageArray = morphology.dilation(imageArray,morphology.disk(3))
return imageArray
def find_mask(base_name, base_img, img_name, img, model_robust, channel):
# what type of interpolation
# 0: nearest-neighbor
# 1: bi-linear
warp_order = 1
output_shape, corner_min = find_output_shape(base_img, model_robust, channel)
# This in-plane offset is the only necessary transformation for the base image
offset = SimilarityTransform(translation= -corner_min)
base_warped = warp(base_img[:,:,channel], offset.inverse, order=warp_order,
output_shape = output_shape, cval=-1)
base_color = warp(base_img, offset.inverse, order=warp_order,
output_shape = output_shape, cval=-1)
# warp image corners to new position in mosaic
transform = (model_robust + offset).inverse
img_warped = warp(img[:,:,channel], transform, order=warp_order,
output_shape=output_shape, cval=-1)
img_color = warp(img, transform, order=warp_order,
output_shape=output_shape, cval=-1)
base_mask = (base_warped != -1)
base_warped[~base_mask] = 0
img_mask = (img_warped != -1)
img_warped[~img_mask] = 0
plt.imsave("img_mask.jpg", img_mask)
#convert to rgb
img_alpha = np.dstack((img_color, img_mask))
base_alpha = np.dstack((base_color, base_mask))
td = config.tmp_dir
tmp_base = os.path.join(td, 'tmp_' + '.'.join(base_name.split('.')[:-1]) + '.png')
tmp_img = os.path.join(td, 'tmp_' + '.'.join(img_name.split('.')[:-1]) + '.png')
tmp_out = os.path.join(td, 'tmp_out_' + '.'.join(base_name.split('.')[:-1]) + '.png')
plt.imsave(tmp_base, base_alpha)
plt.imsave(tmp_img, img_alpha)
cmd = ['enblend', tmp_base, tmp_img, '-o', tmp_out]
p = Popen(cmd, stdin=PIPE, stdout=PIPE, stderr=PIPE)
output, err = p.communicate(b"input data that is passed to subprocess' stdin")
rc = p.returncode
#if you don't have enblend, you can use one of these
#merged_img = simple_merge(base_warped, img_warped, base_mask, img_mask)
#merged_img = minimum_cost_merge(base_warped, img_warped, base_mask, img_mask)
#merged_edges = remove_empty_edges(merged_img)
# remove alpha channel
if os.path.exists(tmp_out):
out = imread(tmp_out)
oute = remove_empty_alpha(out)
os.remove(tmp_base)
os.remove(tmp_img)
os.remove(tmp_out)
return oute[:,:,:3]
else:
print("Could not find out", tmp_out, rc)
raise Exception("failed cmd %s" %cmd)
def skimage_filter_technique(image_path):
img2 = data.imread(image_path, True)
tv_filter = filter.denoise_tv_chambolle(img2, weight=0.1)
cv2.imshow('Gray Scale', tv_filter)
cv2.waitKey(Delay)
return tv_filter
def load_image(path, as_grey = False, to_float = True):
# Load image
image = imread(path, as_grey)
if to_float:
# Convert to floating point matrix
image = image.astype(np.float32)
return image