def smooth_temperatures(self, method='mean', window=3):
if method == 'mean':
self.T1_smooth = filters.mean_filter(self.T1, window)
self.T2_smooth = filters.mean_filter(self.T2, window)
elif method == 'median':
self.T1_smooth = filters.median_filter(self.T1, window)
self.T2_smooth = filters.median_filter(self.T2, window)
else:
raise ValueError("Unsupported method. Use only 'mean' or 'median'")
self.T1_res = self.T1 - self.T1_smooth
self.T2_res = self.T2 - self.T2_smooth
def main():
parser = build_parser()
args = vars(parser.parse_args(sys.argv[1:]))
min_range = args['min_range']
max_range = args['max_range']
measurement_interval = args["measurement_interval"]
scan_rate = args["scan_rate"]
seed = args["seed"]
num_measurements = args["num_measurements"]
num_previous_scans = args["D"]
transform = compose(range_filter(min_range, max_range),
median_filter(num_measurements, num_previous_scans))
queue = Queue()
t1 = LidarDriver(queue,
scan_rate=scan_rate,
num_measurements=num_measurements,
rand_interval=measurement_interval,
seed=seed)
t1.daemon = True
t1.start()
process_events(queue,
reactive_transduce(transducer=transform, target=rprint()))
def generate_median_filter_input():
interface = InterfaceInfo.get_instance()
if interface.current_image is not None:
interface.delete_widgets(interface.buttons_frame)
ttk.Label(interface.buttons_frame, text="Windows size", background=constants.TOP_COLOR).grid(row=0, column=0)
windows_size = Entry(interface.buttons_frame)
windows_size.grid(row=0, column=1)
generate_noise = ttk.Button(interface.buttons_frame, text="Apply",
command=lambda: median_filter(interface.current_image, constants.WIDTH,
constants.HEIGHT, int(windows_size.get())))
generate_noise.grid(row=1, column=0)
else:
interface.reset_parameters()
messagebox.showerror(title="Error", message="You must upload an image to apply median filter ")
def all_filters(frame,scale_percent = 20, threshVal1 = 120, threshVal2 = 255,thresholdVal1_inv = 70,thresholdVal2_inv = 255,
thresholdVal3 = 0, thresholdVal4 = 100,median_filter_size = 5,blackign =3):
# ret, frame = cap.read()
# print (frame.shape)
width = int(frame.shape[1] * scale_percent / 100)
height = int(frame.shape[0] * scale_percent / 100)
dsize = (width, height)
frame = cv2.resize(frame, dsize)
H = cv2.cvtColor(frame,cv2.COLOR_RGB2HSV_FULL)
# H = cv2.cvtColor(frame,cv2.COLOR_RGB2GRAY)
# print (frame.shape)
frame2 = threshold(H[:,:,1].copy(),threshVal1,threshVal2)
frame3 = threshold_inv(frame[:,:,1].copy(),thresholdVal1_inv,thresholdVal2_inv)
frame4 = threshold(frame[:,:,0].copy(),thresholdVal3,thresholdVal4)
frame2 = np.array(frame2,np.uint8)
frame3 = np.array(frame3,np.uint8)
frame4 = np.array(frame4,np.uint8)
# print (frame4.shape)
frame2 = median_filter(frame2,filter_size = median_filter_size,plot=False)
frame3 = median_filter(frame3,filter_size = median_filter_size,plot=False)
frame4 = median_filter(frame4,filter_size = median_filter_size,plot=False)
# print (frame4.shape)
frame3 = blacker(frame3,blackign)
# frame3,a = sobel(frame3,True)
frame3 = gaussian(frame3.copy(),9,plot=False)
# frame4 = laplacian(frame2)
a = pixel_count(frame3)
b = pixel_count(frame4)
if b !=0:
ratio_full = (a/b+1)*100
# print ('ammatasiri methnata wenakan enawa')
return frame2, frame3, frame4, a,b, ratio_full
cols = img.shape[1]
for i in range(0, rows):
for j in range(0, cols):
rand_value = random.random()
if rand_value < p:
new_value = 0
elif rand_value > 1 - p:
new_value = 255
else:
new_value = new_image[i][j]
new_image[i][j] = new_value
return new_image
noised_image = apply_salt_and_pepper(img, 0.3)
denoised_image = median_filter(noised_image, 5)
cv2.imshow('image',noised_image)
cv2.waitKey(0)
cv2.destroyAllWindows()
cv2.imshow('median', denoised_image)
cv2.waitKey(0)
cv2.destroyAllWindows()
width = int(frame.shape[1] * scale_percent / 100)
height = int(frame.shape[0] * scale_percent / 100)
dsize = (width, height)
frame = cv2.resize(frame, dsize)
H = cv2.cvtColor(frame, cv2.COLOR_RGB2HSV_FULL)
# H = cv2.cvtColor(frame,cv2.COLOR_RGB2GRAY)
frame2 = threshold(H[:, :, 1].copy(), 120, 255)
frame3 = threshold_inv(frame[:, :, 1].copy(), 70, 255)
frame4 = threshold(frame[:, :, 0].copy(), 0, 100)
frame2 = np.array(frame2, np.uint8)
frame3 = np.array(frame3, np.uint8)
frame4 = np.array(frame4, np.uint8)
frame2 = median_filter(frame2, filter_size=3, plot=False)
frame3 = median_filter(frame3, filter_size=3, plot=False)
frame4 = median_filter(frame4, filter_size=5, plot=False)
frame3 = blacker(frame3, 300)
# frame3,a = sobel(frame3,True)
frame3 = gaussian(frame3.copy(), 9, plot=False)
# frame4 = laplacian(frame2)
a = pixel_count(frame3)
b = pixel_count(frame4)
# print('a',a,'\nb',b)
ratio_full = (a / b) * 100
if ratio_full >= 80:
print('ratio_full', ratio_full, '%')
"Fig0512(b)(ckt-uniform-plus-saltpepr-prob-pt1).tif")
img2 = skimage.img_as_float(img2)
f, axs = plt.subplots(2, 3, figsize=(12, 9))
disable_tick(axs)
axs[0, 0].imshow(img1, cmap="gray")
axs[0, 0].set_title("uniform noise")
axs[0, 1].imshow(img2, cmap="gray")
axs[0, 1].set_title("add salt & pepper noise")
img3 = F.arithmetic_mean_filter(img2, 5, 5)
axs[0, 2].imshow(img3, cmap="gray")
axs[0, 2].set_title("arithmetic mean filter")
img4 = F.geometric_mean_filter(img2, 5, 5)
axs[1, 0].imshow(img4, cmap="gray")
axs[1, 0].set_title("geometric mean filter")
img5 = F.median_filter(img2, 5, 5)
axs[1, 1].imshow(img5, cmap="gray")
axs[1, 1].set_title("median filter")
img6 = F.alpha_trimmed_filter(img2, 5, 5, 5)
axs[1, 2].imshow(img6, cmap="gray")
axs[1, 2].set_title("alpha trimmed filter")
plt.savefig("output/" + sys.argv[0][:-2] + "png")
plt.show()
def process(img, controls, camera, debug):
global rgb_window_active, hsv_window_active
profile = controls.profile
FRAME_WIDTH = camera.FRAME_WIDTH
FRAME_HEIGHT = camera.FRAME_HEIGHT
img = filters.resize(img, camera.FRAME_WIDTH, camera.FRAME_HEIGHT)
original_img = img
rgb_mask = filters.rgb_threshold(img, profile)
if controls.show_rgb_mask:
cv2.imshow("rgb mask", rgb_mask)
rgb_window_active = True
elif rgb_window_active:
cv2.destroyWindow('rgb mask')
rgb_window_active = False
img = filters.apply_mask(img, rgb_mask)
img = filters.hsv_threshold(img, profile)
if controls.show_hsv_mask:
cv2.imshow("hsv mask", img)
hsv_window_active = True
elif hsv_window_active:
cv2.destroyWindow('hsv mask')
hsv_window_active = False
img = filters.median_filter(img)
if debug:
cv2.imshow('median filter', img)
_, contours, hierarchy = cv2.findContours(img, cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
contour_list = []
for (index, contour) in enumerate(contours):
area = cv2.contourArea(contour)
# peri = cv2.arcLength(contour, True)
# approx = cv2.approxPolyDP(contour, 0.04 * peri, True)
#
# limit the number of contours to process
#
if area > MIN_SQUARE_AREA:
contour_list.append(contour)
color = colors.random()
x, y, w, h = cv2.boundingRect(contour)
#
# if it is a cube, then outbound rectangle should be close to a square
#
if is_not_square(w, h):
cv2.rectangle(original_img, (x, y), (x + w, y + h),
colors.WHITE, 2)
continue
cv2.rectangle(original_img, (x, y), (x + w, y + h), colors.GREEN,
2)
#
# print 'square: %s,%s' % (w,h)
# print w/h, h/w
center_mass_x = x + w / 2
center_mass_y = y + h / 2
angle = get_angle(camera, center_mass_x, center_mass_y)
distance = get_distance(w, CUBE_LENGTH, camera.FOCAL_LENGTH)
# print 'x:%s, y:%s angle:%s ' % ( center_mass_x, center_mass_y, angle )
cv2.drawContours(original_img, contours, index, color, 2)
cv2.circle(original_img, (center_mass_x, center_mass_y), 5, color,
-1)
cv2.line(original_img, (FRAME_WIDTH / 2, FRAME_HEIGHT),
(center_mass_x, center_mass_y), color, 2)
font = cv2.FONT_HERSHEY_DUPLEX
coordinate_text = 'x:%s y:%s ' % (center_mass_x, center_mass_y)
area_text = 'area:%s width:%s' % (area, w)
angle_text = 'angle:%.2f distance:%s' % (angle, distance)
cv2.putText(original_img, coordinate_text, (x, y - 35), font, .4,
colors.WHITE, 1, cv2.LINE_AA)
cv2.putText(original_img, area_text, (x, y - 20), font, .4,
colors.WHITE, 1, cv2.LINE_AA)
cv2.putText(original_img, angle_text, (x, y - 5), font, .4,
colors.WHITE, 1, cv2.LINE_AA)
# create a line for the center of frame
cv2.line(original_img, (FRAME_WIDTH / 2, FRAME_HEIGHT),
(FRAME_WIDTH / 2, 0), colors.WHITE, 4)
return original_img
def task_2():
img = Image.open('./task_2.png').convert('L')
# 高斯噪声实验
noise_generator.Gaussian_noise(img, 0, 40).save(
'./output/task_2/p1/add_Gaussian_noise.png')
img1 = Image.open('./output/task_2/p1/add_Gaussian_noise.png').convert('L')
# 算术均值
filters.arithmetic_mean_filter(img1, (3, 3)).save(
'./output/task_2/p1/算术平均3x3.png')
filters.arithmetic_mean_filter(img1, (5, 5)).save(
'./output/task_2/p1/算术平均5x5.png')
filters.arithmetic_mean_filter(img1, (7, 7)).save(
'./output/task_2/p1/算术平均7x7.png')
filters.arithmetic_mean_filter(img1, (9, 9)).save(
'./output/task_2/p1/算术平均9x9.png')
# 几何均值
filters.geometric_mean_filter(img1, (3, 3)).save(
'./output/task_2/p1/几何平均3x3.png')
filters.geometric_mean_filter(img1, (5, 5)).save(
'./output/task_2/p1/几何平均5x5.png')
# 中值滤波
filters.median_filter(img1, (3, 3)).save(
'./output/task_2/p1/中值滤波3x3.png')
filters.median_filter(img1, (5, 5)).save(
'./output/task_2/p1/中值滤波5x5.png')
# 盐噪声实验
noise_generator.Salt_and_pepper_noise(img, ps=0.2, pp=0).save(
'./output/task_2/p2/add_Salt_noise.png')
img2 = Image.open('./output/task_2/p2/add_Salt_noise.png')
# 调和均值处理
filters.harmonic_mean_filter(img2, (3, 3)).save(
'./output/task_2/p2/调和均值3x3.png')
filters.harmonic_mean_filter(img2, (5, 5)).save(
'./output/task_2/p2/调和均值5x5.png')
# 逆谐波均值处理 Q > 0
filters.contra_harmonic_mean_filter(img2, (3, 3), Q=1.5).save(
'./output/task_2/p2/逆谐波均值3x3q1.5.png')
# 逆谐波均值处理 Q < 0
filters.contra_harmonic_mean_filter(img2, (3, 3), Q=-1.5).save(
'./output/task_2/p2/逆谐波均值3x3q-1.5.png')
filters.contra_harmonic_mean_filter(img2, (3, 3), Q=-2).save(
'./output/task_2/p2/逆谐波均值3x3q-2.0.png')
filters.contra_harmonic_mean_filter(img2, (3, 3), Q=-2.5).save(
'./output/task_2/p2/逆谐波均值3x3q-2.5.png')
# 椒盐噪声实验
noise_generator.Salt_and_pepper_noise(img, ps=0.2, pp=0.2).save(
'./output/task_2/p3/add_Salt_and_pepper_noise.png')
img3 = Image.open('./output/task_2/p3/add_Salt_and_pepper_noise.png')
# 算术均值
filters.arithmetic_mean_filter(img3, (3, 3)).save(
'./output/task_2/p3/算术平均3x3.png')
filters.arithmetic_mean_filter(img3, (5, 5)).save(
'./output/task_2/p3/算术平均5x5.png')
# 几何均值
filters.geometric_mean_filter(img3, (3, 3)).save(
'./output/task_2/p3/几何平均3x3.png')
filters.geometric_mean_filter(img3, (5, 5)).save(
'./output/task_2/p3/几何平均5x5.png')
# 最大值滤波
filters.max_filter(img3, (3, 3)).save(
'./output/task_2/p3/最大值滤波3x3.png')
filters.max_filter(img3, (5, 5)).save(
'./output/task_2/p3/最大值滤波5x5.png')
# 最小值滤波
filters.min_filter(img3, (3, 3)).save(
'./output/task_2/p3/最小值滤波3x3.png')
filters.min_filter(img3, (5, 5)).save(
'./output/task_2/p3/最小值滤波5x5.png')
# 中值滤波
filters.median_filter(img3, (3, 3)).save(
'./output/task_2/p3/中值滤波3x3.png')
filters.median_filter(img3, (5, 5)).save(
'./output/task_2/p3/中值滤波5x5.png')
def disable_tick(axs):
for u in range(len(axs)):
axs[u].axes.get_xaxis().set_ticks([])
axs[u].axes.get_yaxis().set_ticks([])
img1 = io.imread("input/"+"Fig0514(a)(ckt_saltpep_prob_pt25).tif")
img1 = skimage.img_as_float(img1)
f,axs = plt.subplots(1,3,figsize=(12,9))
disable_tick(axs)
axs[0].imshow(img1,cmap="gray")
axs[0].set_title("salt & pepper noise")
img2 = F.median_filter(img1,7,7)
axs[1].imshow(img2,cmap="gray")
axs[1].set_title("median filter 7x7")
img3 = F.adaptive_median_filter(img1,3)
axs[2].imshow(img3,cmap="gray")
axs[2].set_title("adaptive median filter")
plt.savefig("output/"+sys.argv[0][:-2]+"png")
plt.show()
def setUp(self):
self.res = []
transducer = compose(range_filter(START, END), median_filter(N, D))
self.target = reactive_transduce(transducer=transducer,
target=list_builder(self.res))
def setUp(self):
self.res = []
transducer = median_filter(N, D)
self.target = reactive_transduce(transducer=transducer,
target=list_builder(self.res))
for u in range(len(axs)):
for v in range(len(axs[0])):
axs[u, v].axes.get_xaxis().set_ticks([])
axs[u, v].axes.get_yaxis().set_ticks([])
img1 = io.imread("input/" + "Fig0510(a)(ckt-board-saltpep-prob.pt05).tif")
img1 = skimage.img_as_float(img1)
f, axs = plt.subplots(2, 3, figsize=(12, 9))
disable_tick(axs)
axs[0, 0].imshow(img1, cmap="gray")
axs[0, 0].set_title("salt & pepper noise p=0.1")
img2 = F.median_filter(img1, 3, 3)
axs[0, 1].imshow(img2, cmap="gray")
axs[0, 1].set_title("median filter 1 times")
img3 = F.median_filter(img2, 3, 3)
axs[0, 2].imshow(img3, cmap="gray")
axs[0, 2].set_title("median filter 2 times")
img4 = F.median_filter(img1, 3, 3)
axs[1, 0].imshow(img4, cmap="gray")
axs[1, 0].set_title("median filter 3 times")
img5 = io.imread("input/Fig0508(a)(circuit-board-pepper-prob-pt1).tif")
img5 = F.max_filter(img5, 3, 3)
axs[1, 1].imshow(img5, cmap="gray")
axs[1, 1].set_title("max filter")
