from global_segmentation import (
read_pgm, )
GT = read_pgm(f'Data\\Ground Truth\\TEST_55.pgm')
IM = read_pgm(f'Data\\Brain Volume\\TEST_55.pgm')
Y = 217 # height
X = 181 # col
intensity = {}
for k in range(Y):
for l in range(X):
if GT[k][l] not in intensity:
intensity[GT[k][l]] = [IM[k][l]]
else:
intensity[GT[k][l]].append(IM[k][l])
for ele in intensity:
print(ele, " ", sum(intensity[ele]) / len(intensity[ele]))
N_IT = 100
C = [13,45,100,115] # Values of the initial Clusters
A = 0.7 # Value of alpha
N = len(C)
P = 1
Q = 3
M = 2.5
E = 0.01
p = (1/(M-1))
img_vol = []
ground_truth = []
for j in range(START,END):
img_vol.append(read_pgm(f'Data\\Brain Volume\\TEST_{j}.pgm'))
ground_truth.append(read_pgm(f'Data\\Ground Truth\\TEST_{j}.pgm'))
print("Z = ",len(img_vol))
print("Y = ",len(img_vol[0]))
print("X = ",len(img_vol[0][0]))
global_u = full((len(C),Z,Y,X),0).tolist()
local_u = full((len(C),Z,Y,X),0).tolist()
final_u = full((len(C),Z,Y,X),0).tolist()
for i in range(N):
for j in range(Z):
for k in range(Y):
for l in range(X):
def intersection(lst1, lst2):
return list(set(lst1) & set(lst2))
START = 50
END = 52
Z = (END - START) # depth
Y = 217 # height
X = 181 # col
test_img = []
ground_truth = []
for j in range(START, END):
test_img.append(read_pgm(f'Data\\Test Results\\TEST_{j}.pgm'))
ground_truth.append(read_pgm(f'Data\\Ground Truth\\TEST_{j}.pgm'))
correct_classified = 0
mis_classified = 0
# Misclassification Error
for j in range(Z):
for k in range(Y):
for l in range(X):
if ground_truth[j][k][l] in [0, 1, 2, 3]:
if ground_truth[j][k][l] == test_img[j][k][l]:
correct_classified += 1
else:
mis_classified += 1