def gaussianlowpass(f, imgs):
r, c = imgs[:2]
ct = (r / 2, c / 2)
b = np.zeros(imgs[:2])
for x in range(c):
for y in range(r):
b[y, x] = math.exp(((-dist((y, x), ct)**2) / (2 * (f**2))))
return b
def butterlowpass(f, imgs, n):
r, c = imgs[:2]
ct = (r / 2, c / 2)
b = np.zeros(imgs[:2])
for x in range(c):
for y in range(r):
b[y, x] = 1 / (1 + (dist((y, x), ct) / f)**(2 * n))
return b
def idealhighpass(f, imgs):
r, c = imgs[:2]
ct = (r / 2, c / 2)
b = np.ones(imgs[:2])
for x in range(c):
for y in range(r):
if dist((y, x), ct) < f:
b[y, x] = 0
return b
def ideallowpass(f, imgs):
r, c = imgs[:2]
ct = (r / 2, c / 2)
b = np.zeros(imgs[:2])
for x in range(c):
for y in range(r):
if dist((y, x), ct) < f:
b[y, x] = 1
return b
