def fade_scores():
i = 0
for filename in os.listdir(with_coloration_path):
# print("Filename: ", filename)
# print("Inside first for loop")
i += 1
im_path = with_coloration_path + filename
inp_img = cv2.imread(im_path)
fade_inp_img = compute_fade(inp_img)
predicted_path = get_path(inp_img, 117) # threshold=117
ws['A{}'.format(i + 1)] = filename
ws['B{}'.format(i + 1)] = fade_inp_img
dehazenet_image = cv2.imread(
dehazenet_coloration_path +
'{}_finalWithoutCLAHE.jpg'.format(filename[:-4]))
fade_dehazenet = compute_fade(dehazenet_image)
ws['C{}'.format(i + 1)] = fade_dehazenet
if predicted_path == 1:
enhanced_img = apply_CLAHE(adaptive_enhance(inp_img))
ws['D{}'.format(i + 1)] = compute_fade(enhanced_img)
else:
ws['D{}'.format(i + 1)] = fade_dehazenet
# Comparison with raw input image scores
# In FADE, lesser score is better
if ws['D{}'.format(i + 1)].value < ws['B{}'.format(i + 1)].value:
ws['E{}'.format(i + 1)] = "Yes"
elif ws['D{}'.format(i + 1)].value > ws['B{}'.format(i + 1)].value:
ws['E{}'.format(i + 1)] = "No"
else:
ws['E{}'.format(i + 1)] = "Equal"
# Comparison with DehazeNet scores
if ws['D{}'.format(i + 1)].value < ws['C{}'.format(i + 1)].value:
ws['F{}'.format(i + 1)] = "Yes"
elif ws['D{}'.format(i + 1)].value > ws['C{}'.format(i + 1)].value:
ws['F{}'.format(i + 1)] = "No"
else:
ws['F{}'.format(i + 1)] = "Equal"
# ws['D{}'.format(i + 1)] = niqe(dehaze_image(inp_img, im_path))
# ws['C{}'.format(i + 1)] = ws['D{}'.format(i + 1)].value
print("FADE computed for: ", filename)
for filename in os.listdir(without_coloration_path):
# print("Filename: ", filename)
# print("Inside first for loop")
i += 1
im_path = without_coloration_path + filename
inp_img = cv2.imread(im_path)
fade_inp_img = compute_fade(inp_img)
predicted_path = get_path(inp_img, 117) # threshold=117
ws['A{}'.format(i + 1)] = filename
ws['B{}'.format(i + 1)] = fade_inp_img
dehazenet_image = cv2.imread(
dehazenet_without_coloration_path +
'{}_finalWithoutCLAHE.jpg'.format(filename[:-4]))
fade_dehazenet = compute_fade(dehazenet_image)
ws['C{}'.format(i + 1)] = fade_dehazenet
if predicted_path == 1:
enhanced_img = apply_CLAHE(adaptive_enhance(inp_img))
ws['D{}'.format(i + 1)] = compute_fade(enhanced_img)
else:
ws['D{}'.format(i + 1)] = fade_dehazenet
# Comparison with raw input image scores
if ws['D{}'.format(i + 1)].value < ws['B{}'.format(i + 1)].value:
ws['E{}'.format(i + 1)] = "Yes"
elif ws['D{}'.format(i + 1)].value > ws['B{}'.format(i + 1)].value:
ws['E{}'.format(i + 1)] = "No"
else:
ws['E{}'.format(i + 1)] = "Equal"
# Comparison with DehazeNet scores
if ws['D{}'.format(i + 1)].value < ws['C{}'.format(i + 1)].value:
ws['F{}'.format(i + 1)] = "Yes"
elif ws['D{}'.format(i + 1)].value > ws['C{}'.format(i + 1)].value:
ws['F{}'.format(i + 1)] = "No"
else:
ws['F{}'.format(i + 1)] = "Equal"
# op_img = dehaze_image(inp_img, im_path)
# niqe_score = niqe(op_img)
# ws['D{}'.format(i + 1)] = niqe_score
# ws['C{}'.format(i + 1)] = ws['D{}'.format(i + 1)].value
print("FADE computed for: ", filename)
wb.save(excel_path)
print("Script complete and excel saved!")
def ssim_scores():
i = 0
for filename in os.listdir(with_coloration_path):
i += 1
gt_name = filename[:filename.find(
'_')] + '.png' # till the first occurence of underscore
# print(gt_name)
# print(gt_path + gt_name)
gt_img = cv2.imread(gt_path + gt_name)
im_path = with_coloration_path + filename
inp_img = cv2.imread(im_path)
predicted_path = get_path(inp_img, 117) # threshold=117
ws['A{}'.format(i + 1)] = filename
ws['B{}'.format(i + 1)] = structural_similarity(gt_img,
inp_img,
multichannel=True)
dehazenet_image = cv2.imread(
dehazenet_coloration_path +
'{}_finalWithoutCLAHE.jpg'.format(filename[:-4]))
ssim_dehaze = structural_similarity(gt_img,
dehazenet_image,
multichannel=True)
ws['C{}'.format(i + 1)] = ssim_dehaze
if predicted_path == 1:
ws['D{}'.format(i + 1)] = structural_similarity(
gt_img,
apply_CLAHE(adaptive_enhance(inp_img)),
multichannel=True)
else:
ws['D{}'.format(i + 1)] = ssim_dehaze
# Comparison with raw input image scores
if ws['D{}'.format(i + 1)].value > ws['B{}'.format(i + 1)].value:
ws['E{}'.format(i + 1)] = "Yes"
elif ws['D{}'.format(i + 1)].value < ws['B{}'.format(i + 1)].value:
ws['E{}'.format(i + 1)] = "No"
else:
ws['E{}'.format(i + 1)] = "Equal"
# Comparison with DehazeNet scores
if ws['D{}'.format(i + 1)].value > ws['C{}'.format(i + 1)].value:
ws['F{}'.format(i + 1)] = "Yes"
elif ws['D{}'.format(i + 1)].value < ws['C{}'.format(i + 1)].value:
ws['F{}'.format(i + 1)] = "No"
else:
ws['F{}'.format(i + 1)] = "Equal"
print("SSIM computed for: ", filename)
for filename in os.listdir(without_coloration_path):
i += 1
gt_name = filename[:filename.find(
'_')] + '.png' # till the first occurence of underscore
gt_img = cv2.imread(gt_path + gt_name)
im_path = without_coloration_path + filename
inp_img = cv2.imread(im_path)
predicted_path = get_path(inp_img, 117) # threshold=117
ws['A{}'.format(i + 1)] = filename
ws['B{}'.format(i + 1)] = structural_similarity(gt_img,
inp_img,
multichannel=True)
dehazenet_image = cv2.imread(
dehazenet_without_coloration_path +
'{}_finalWithoutCLAHE.jpg'.format(filename[:-4]))
ssim_dehaze = structural_similarity(gt_img,
dehazenet_image,
multichannel=True)
ws['C{}'.format(i + 1)] = ssim_dehaze
if predicted_path == 1:
ws['D{}'.format(i + 1)] = structural_similarity(
gt_img,
apply_CLAHE(adaptive_enhance(inp_img)),
multichannel=True)
else:
ws['D{}'.format(i + 1)] = ssim_dehaze
# Comparison with raw input image scores
if ws['D{}'.format(i + 1)].value > ws['B{}'.format(i + 1)].value:
ws['E{}'.format(i + 1)] = "Yes"
elif ws['D{}'.format(i + 1)].value < ws['B{}'.format(i + 1)].value:
ws['E{}'.format(i + 1)] = "No"
else:
ws['E{}'.format(i + 1)] = "Equal"
# Comparison with DehazeNet scores
if ws['D{}'.format(i + 1)].value > ws['C{}'.format(i + 1)].value:
ws['F{}'.format(i + 1)] = "Yes"
elif ws['D{}'.format(i + 1)].value < ws['C{}'.format(i + 1)].value:
ws['F{}'.format(i + 1)] = "No"
else:
ws['F{}'.format(i + 1)] = "Equal"
print("SSIM computed for: ", filename)
wb.save(excel_path)
print("Script complete and excel saved!")
I = src / 255.0
dark = DarkChannel(I, 15)
A = AtmLight(I, dark)
te = TransmissionEstimate(im_path, height, width)
t = TransmissionRefine(src, te)
J = Recover(I, t, A, 0.1)
dehazed_image = J * 255
return dehazed_image
if __name__ == '__main__':
if not len(sys.argv) == 2:
print('Usage: python DeHazeNet.py haze_img_path')
exit()
else:
im_path = sys.argv[1]
src = cv2.imread(im_path)
path = get_path(src, threshold=137)
if path == 1:
dehazed_image = apply_CLAHE(src)
else:
dehazed_image = dehaze_image(src, im_path)
# cv2.imshow('TransmissionEstimate', te)
# cv2.imshow('TransmissionRefine', t)
# cv2.imshow('Origin', src)
# cv2.imshow('Dehaze', J)
# cv2.waitKey(0)
save_path = im_path[:-4] + '_Dehaze' + im_path[-4:len(im_path)]
cv2.imwrite(save_path, dehazed_image)
print("Dehazed image saved!")