def Reputation_threshold(image1, image2, image3, image4, image5,
crowd_density_array):
input_image1 = cv2.imread(str(image1))
input_image2 = cv2.imread(str(image2))
input_image3 = cv2.imread(str(image3))
input_image4 = cv2.imread(str(image4))
input_image5 = cv2.imread(str(image5))
input_image_1 = fix_image_size(input_image1)
blur_map_1, score_1, blurry_1 = estimate_blur(input_image_1)
input_image_2 = fix_image_size(input_image2)
blur_map_2, score_2, blurry_2 = estimate_blur(input_image_2)
input_image_3 = fix_image_size(input_image3)
blur_map_3, score_3, blurry_3 = estimate_blur(input_image_3)
input_image_4 = fix_image_size(input_image4)
blur_map_4, score_4, blurry_4 = estimate_blur(input_image_4)
input_image_5 = fix_image_size(input_image5)
blur_map_5, score_5, blurry_5 = estimate_blur(input_image_5)
# print("Quality_score1: {0}, blurry1: {1}".format(score_1, blurry_1))
# print("Quality_score2: {0}, blurry2: {1}".format(score_2, blurry_2))
# print("Quality_score3: {0}, blurry3: {1}".format(score_3, blurry_3))
# print("Quality_score4: {0}, blurry4: {1}".format(score_4, blurry_4))
# print("Quality_score5: {0}, blurry5: {1}".format(score_5, blurry_5))
""" if args.display:
cv2.imshow("input", input_image1)
cv2.imshow("result", pretty_blur_map(blur_map))
cv2.waitKey(0)"""
img_gray = cv2.cvtColor(input_image1, cv2.COLOR_BGR2GRAY)
noise_value_img_1 = estimate_noise(img_gray)
img_gray = cv2.cvtColor(input_image2, cv2.COLOR_BGR2GRAY)
noise_value_img_2 = estimate_noise(img_gray)
img_gray = cv2.cvtColor(input_image3, cv2.COLOR_BGR2GRAY)
noise_value_img_3 = estimate_noise(img_gray)
img_gray = cv2.cvtColor(input_image4, cv2.COLOR_BGR2GRAY)
noise_value_img_4 = estimate_noise(img_gray)
img_gray = cv2.cvtColor(input_image5, cv2.COLOR_BGR2GRAY)
noise_value_img_5 = estimate_noise(img_gray)
# print("Noise of camera 1 is ", noise_value_img_1)
# print("Noise of camera 2 is ", noise_value_img_2)
# print("Noise of camera 3 is ", noise_value_img_3)
# print("Noise of camera 4 is ", noise_value_img_4)
# print("Noise of camera 5 is ", noise_value_img_5)
global overlapping_array
view_7 = cameras(0, overlapping_array)
view_1 = cameras(1, [0.3, 1, 0.3, 0.5, 0.4])
view_5 = cameras(2, [0.5, 0.3, 1, 0.4, 0.4])
view_6 = cameras(3, [0.4, 0.5, 0.4, 1, 0.4])
view_8 = cameras(4, [0.2, 0.5, 0.3, 0.4, 1])
# crowd density will be calculated from a separate code.These are dummy values
#scalar = StandardScaler(with_mean=False)
#crowd_density_array = [[10], [5], [7]]
crowd_density_array = [crowd_density_array]
#scalar.fit(crowd_density_array)
#crowd_density_array = scalar.transform(crowd_density_array)
#[crowd_density_array] = np.array(crowd_density_array).reshape((1, 3))
[crowd_density_array] = normalize(crowd_density_array, norm='l2')
view_7.crowd_density([view_7.cid, crowd_density_array[view_7.cid]])
view_7.crowd_density([view_1.cid, crowd_density_array[view_1.cid]])
view_7.crowd_density([view_5.cid, crowd_density_array[view_5.cid]])
view_7.crowd_density([view_6.cid, crowd_density_array[view_6.cid]])
view_7.crowd_density([view_8.cid, crowd_density_array[view_8.cid]])
view_1.crowd_density([view_1.cid, crowd_density_array[view_1.cid]])
view_1.crowd_density([view_7.cid, crowd_density_array[view_7.cid]])
view_1.crowd_density([view_5.cid, crowd_density_array[view_5.cid]])
view_1.crowd_density([view_6.cid, crowd_density_array[view_6.cid]])
view_1.crowd_density([view_8.cid, crowd_density_array[view_8.cid]])
view_5.crowd_density([view_5.cid, crowd_density_array[view_5.cid]])
view_5.crowd_density([view_7.cid, crowd_density_array[view_7.cid]])
view_5.crowd_density([view_1.cid, crowd_density_array[view_1.cid]])
view_5.crowd_density([view_6.cid, crowd_density_array[view_6.cid]])
view_5.crowd_density([view_8.cid, crowd_density_array[view_8.cid]])
#scalar = StandardScaler(with_mean=False)
#noise_array = [[noise_value_img_1], [noise_value_img_2], [noise_value_img_3]]
noise_array = [[
noise_value_img_1, noise_value_img_2, noise_value_img_3,
noise_value_img_4, noise_value_img_5
]]
#scalar.fit(noise_array)
#noise_array = scalar.transform(noise_array)
#[noise_array] = np.array(noise_array).reshape((1, 3))
[noise_array] = normalize(noise_array, norm='l2')
view_7.noise([view_7.cid, noise_array[view_7.cid]])
view_7.noise([view_1.cid, noise_array[view_1.cid]])
view_7.noise([view_5.cid, noise_array[view_5.cid]])
view_7.noise([view_6.cid, noise_array[view_6.cid]])
view_7.noise([view_8.cid, noise_array[view_8.cid]])
view_1.noise([view_7.cid, noise_array[view_7.cid]])
view_1.noise([view_1.cid, noise_array[view_1.cid]])
view_1.noise([view_5.cid, noise_array[view_5.cid]])
view_1.noise([view_6.cid, noise_array[view_6.cid]])
view_1.noise([view_8.cid, noise_array[view_8.cid]])
view_5.noise([view_7.cid, noise_array[view_7.cid]])
view_5.noise([view_1.cid, noise_array[view_1.cid]])
view_5.noise([view_5.cid, noise_array[view_5.cid]])
view_5.noise([view_6.cid, noise_array[view_6.cid]])
view_5.noise([view_8.cid, noise_array[view_8.cid]])
#scalar = StandardScaler(with_mean=False)
#blur_array = [[score_1], [score_2], [score_3]]
blur_array = [[score_1, score_2, score_3, score_4, score_5]]
#scalar.fit(blur_array)
#blur_array = scalar.transform(blur_array)
#[blur_array] = np.array(blur_array).reshape((1, 3))
[blur_array] = normalize(blur_array, norm='l2')
view_7.blur([view_7.cid, blur_array[view_7.cid]])
view_7.blur([view_1.cid, blur_array[view_1.cid]])
view_7.blur([view_5.cid, blur_array[view_5.cid]])
view_7.blur([view_6.cid, blur_array[view_6.cid]])
view_7.blur([view_8.cid, blur_array[view_8.cid]])
view_1.blur([view_1.cid, blur_array[view_1.cid]])
view_1.blur([view_7.cid, blur_array[view_7.cid]])
view_1.blur([view_5.cid, blur_array[view_5.cid]])
view_1.blur([view_6.cid, blur_array[view_6.cid]])
view_1.blur([view_8.cid, blur_array[view_8.cid]])
view_5.blur([view_5.cid, blur_array[view_5.cid]])
view_5.blur([view_7.cid, blur_array[view_7.cid]])
view_5.blur([view_1.cid, blur_array[view_1.cid]])
view_5.blur([view_6.cid, blur_array[view_6.cid]])
view_5.blur([view_8.cid, blur_array[view_8.cid]])
thresh_view_7 = list(view_7.threshold_calc())
#thresh_view_1 = view_1.threshold_calc()
#thresh_view_5 = view_5.threshold_calc()
# print("Threshold of cameras w.r.t to View7 is ", thresh_view_7)
return thresh_view_7
for image_path in find_images(args.images):
image = cv2.imread(str(image_path))
if image is None:
logging.warning(
f'warning! failed to read image from {image_path}; skipping!')
continue
logging.info(f'processing {image_path}')
if fix_size:
image = fix_image_size(image)
else:
logging.warning(
'not normalizing image size for consistent scoring!')
blur_map, score, blurry = estimate_blur(image,
threshold=args.threshold)
logging.info(
f'image_path: {image_path} score: {score} blurry: {blurry}')
results.append({
'input_path': str(image_path),
'score': score,
'blurry': blurry
})
if args.display:
cv2.imshow('input', image)
cv2.imshow('result', pretty_blur_map(blur_map))
if cv2.waitKey(0) == ord('q'):
logging.info('exiting...')
help='set logging level to debug',
action="store_true")
parser.add_argument("-d",
"--display",
dest='display',
help='display images',
action="store_true")
args = parser.parse_args()
if args.verbose:
logging.basicConfig(level=logging.DEBUG)
else:
logging.basicConfig(level=logging.INFO)
assert os.path.exists(args.input_image)
input_image = cv2.imread(args.input_image)
if args.fix_size:
input_image = fix_image_size(input_image)
blur_map, score, blurry = estimate_blur(input_image)
logging.info("score: {0}, blurry: {1}".format(score, blurry))
if args.display:
cv2.imshow("input", input_image)
cv2.imshow("result", pretty_blur_map(blur_map))
cv2.waitKey(0)
self.threshold[count] += weightage[3] * i
count += 1
[self.threshold] = normalize([self.threshold], norm='l2')
return self.threshold
print("Enter the input images name for the three cameras")
s1, s2, s3 = raw_input().split()
input_image1 = cv2.imread(s1)
input_image2 = cv2.imread(s2)
input_image3 = cv2.imread(s3)
input_image_1 = fix_image_size(input_image1)
blur_map_1, score_1, blurry_1 = estimate_blur(input_image_1)
input_image_2 = fix_image_size(input_image2)
blur_map_2, score_2, blurry_2 = estimate_blur(input_image_2)
input_image_3 = fix_image_size(input_image3)
blur_map_3, score_3, blurry_3 = estimate_blur(input_image_3)
print("Quality_score1: {0}, blurry1: {1}".format(score_1, blurry_1))
print("Quality_score2: {0}, blurry2: {1}".format(score_2, blurry_2))
print("Quality_score3: {0}, blurry3: {1}".format(score_3, blurry_3))
""" if args.display:
cv2.imshow("input", input_image1)
cv2.imshow("result", pretty_blur_map(blur_map))
cv2.waitKey(0)"""