def process(nameImage):
# result = fingerprint_pipline('101_3.tif')
output_dir = './output/'
block_size = 16
im = cv.imread('./input/' + nameImage, 0)
normalized_img = normalize(im.copy(), float(100), float(100))
# ROI and normalisation
(segmented_img, normim,
mask) = create_segmented_and_variance_images(normalized_img, block_size,
0.2)
# orientations
angles = orientation.calculate_angles(normalized_img,
W=block_size,
smoth=False)
orientation_img = orientation.visualize_angles(segmented_img,
mask,
angles,
W=block_size)
# find the overall frequency of ridges in Wavelet Domain
freq = ridge_freq(normim,
mask,
angles,
block_size,
kernel_size=5,
minWaveLength=5,
maxWaveLength=15)
# create gabor filter and do the actual filtering
gabor_img = gabor_filter(normim, angles, freq)
# thinning oor skeletonize
thin_image = skeletonize(gabor_img)
# minutias
(minutias, result_minutias) = calculate_minutiaes(thin_image)
# singularities
(singularities_img,
result_singulares) = calculate_singularities(thin_image, angles, 1, 15,
mask)
# visualize pipeline stage by stage
output_imgs = [
im, normalized_img, segmented_img, orientation_img, gabor_img,
thin_image, minutias, singularities_img
]
for i in range(len(output_imgs)):
if len(output_imgs[i].shape) == 2:
output_imgs[i] = cv.cvtColor(output_imgs[i], cv.COLOR_GRAY2RGB)
results = np.concatenate([
np.concatenate(output_imgs[:4], 1),
np.concatenate(output_imgs[4:], 1)
]).astype(np.uint8)
cv.imwrite(output_dir + nameImage + '.png', results)
return result_minutias, result_singulares
def prediction(args):
thetas = load_thetas()
kms, prices = load_dataset()
norm_mileage = normalize(args.mileage, kms)
estimation = estimate_price(*thetas, norm_mileage)
if sum(thetas) != 0:
estimation = denormalize(estimation, prices)
print(f"Estimation for {args.mileage} is {round(estimation, 2)}")
def fingerprint_pipline(input_img):
block_size = 16
# pipe line picture re https://www.cse.iitk.ac.in/users/biometrics/pages/111.JPG
# normalization -> orientation -> frequency -> mask -> filtering
# normalization - removes the effects of sensor noise and finger pressure differences.
normalized_img = normalize(input_img.copy(), float(100), float(100))
# color threshold
# threshold_img = normalized_img
# _, threshold_im = cv.threshold(normalized_img,127,255,cv.THRESH_OTSU)
# cv.imshow('color_threshold', normalized_img); cv.waitKeyEx()
# ROI and normalisation
(segmented_img, normim,
mask) = create_segmented_and_variance_images(normalized_img, block_size,
0.2)
# orientations
angles = orientation.calculate_angles(normalized_img,
W=block_size,
smoth=False)
orientation_img = orientation.visualize_angles(segmented_img,
mask,
angles,
W=block_size)
# find the overall frequency of ridges in Wavelet Domain
freq = ridge_freq(normim,
mask,
angles,
block_size,
kernel_size=5,
minWaveLength=5,
maxWaveLength=15)
# create gabor filter and do the actual filtering
gabor_img = gabor_filter(normim, angles, freq)
# thinning oor skeletonize
thin_image = skeletonize(gabor_img)
# minutias
minutias, pos = calculate_minutiaes(thin_image)
# singularities
#singularities_img = calculate_singularities(thin_image, angles, 1, block_size, mask)
# visualize pipeline stage by stage
return minutias, pos
def fingerprint(input_img):
block_size = 16
# pipe line picture re https://www.cse.iitk.ac.in/users/biometrics/pages/111.JPG
# normalization -> orientation -> frequency -> mask -> filtering
# normalization - removes the effects of sensor noise and finger pressure differences.
normalized_img = normalize(input_img.copy(), float(100), float(100))
# color threshold
# threshold_img = normalized_img
# _, threshold_im = cv.threshold(normalized_img,127,255,cv.THRESH_OTSU)
# cv.imshow('color_threshold', normalized_img); cv.waitKeyEx()
# ROI and normalisation
(segmented_img, normim,
mask) = create_segmented_and_variance_images(normalized_img, block_size,
0.2)
# orientations
angles = orientation.calculate_angles(normalized_img,
W=block_size,
smoth=False)
orientation_img = orientation.visualize_angles(segmented_img,
mask,
angles,
W=block_size)
# find the overall frequency of ridges in Wavelet Domain
freq = ridge_freq(normim,
mask,
angles,
block_size,
kernel_size=5,
minWaveLength=5,
maxWaveLength=15)
# create gabor filter and do the actual filtering
gabor_img = gabor_filter(normim, angles, freq)
# thinning oor skeletonize
thin_image = skeletonize(gabor_img)
# minutias
minutias = calculate_minutiaes(thin_image)
# singularities
singularities_img = calculate_singularities(thin_image, angles, 1,
block_size, mask)
# visualize pipeline stage by stage
output_imgs = [
input_img, normalized_img, segmented_img, orientation_img, gabor_img,
thin_image, minutias, singularities_img
]
for i in range(len(output_imgs)):
if len(output_imgs[i].shape) == 2:
output_imgs[i] = cv.cvtColor(output_imgs[i], cv.COLOR_GRAY2RGB)
results = np.concatenate([
np.concatenate(output_imgs[:4], 1),
np.concatenate(output_imgs[4:], 1)
]).astype(np.uint8)
return results
def _normalize_dataset(kms, prices):
normalized_kms = map(lambda km: normalize(km, kms), kms)
normalized_prices = map(lambda price: normalize(price, prices), prices)
return [Data(x, y) for x, y in zip(normalized_kms, normalized_prices)]