y.append(1 - tmr_arr[it9])
return y
def syndrome_fix(syndrome_vector_noised):
index_to_fix = -1
if np.array_equal(syndrome_vector_noised, np.zeros((4, ), dtype=int)):
return -1
for columnIt in range(0, len(H)):
if np.array_equal(syndrome_vector_noised, H.T[columnIt]):
index_to_fix = columnIt
return index_to_fix
image = cv2.imread("outFile_ITL_Gilbert_2.png", 1)
x = ldpc_images.RGB2Bin(image)
y = np.zeros_like(x)
input_data = np.zeros((4, ), dtype=int)
it3 = 0
main_error_counter = 0
for it in range(len(x)):
for it2 in range(len(x[it])):
it3 = 0
while it3 < len(x[it][it2]):
input_data = np.copy(x[it][it2][it3:it3 + 4])
data_vector = np.dot(input_data,
G) % 2 # Calculated input data vector
noised_data = np.asarray(gilbert_noise(
if current_state == 'G':
if np.random.random() < p_gb: # symulacja zmiany stanu z prawdopodobienstwiem pDZ
current_state = 'B' # zmiana stanu
y[it][it2][it3] = x[it][it2][it3]
elif current_state == 'B':
if np.random.random() < p_bg:
current_state = 'G'
y[it][it2][it3] = 1 - x[it][it2][it3]
it2 += x_iterator
it2 = 0
current_state = 'G'
it += y_iterator
counter += 1
it = 0
if counter % 2 == 0:
x_iterator = x_iterator // 2
else:
y_iterator = y_iterator // 2
return y
image = cv2.imread("inFile.png", 1)
image_arr = ldpc_images.RGB2Bin(image)
hamming_arr = np.zeros_like(image_arr)
# hamming_arr = bsc_noise(image_arr, hamming_arr)
hamming_arr = gilbert_noise(image_arr, hamming_arr)
image_final = ldpc_images.Bin2RGB(hamming_arr)
cv2.imwrite("outFileITL.png", image_final)