1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768697071727374757677787980818283848586 |
- from sklearn.externals import joblib
- import numpy as np
- from ipfml import image_processing
- from ipfml import metrics
- from PIL import Image
- from skimage import color
- import sys, os, getopt
- min_max_file_path = 'fichiersSVD_light/mscn_min_max_values'
- def main():
- if len(sys.argv) <= 1:
- print('Run with default parameters...')
- print('python predict_noisy_image_svd_mscn.py --image path/to/xxxx --interval "0,20" --model path/to/xxxx.joblib --mode ["svdn", "svdne"]')
- sys.exit(2)
- try:
- opts, args = getopt.getopt(sys.argv[1:], "hi:t:m:o", ["help=", "image=", "interval=", "model=", "mode="])
- except getopt.GetoptError:
- # print help information and exit:
- print('python predict_noisy_image_svd_mscn.py --image path/to/xxxx --interval "xx,xx" --model path/to/xxxx.joblib --mode ["svdn", "svdne"]')
- sys.exit(2)
- for o, a in opts:
- if o == "-h":
- print('python predict_nopredict_noisy_image_svd_mscnisy_image.py --image path/to/xxxx --interval "xx,xx" --model path/to/xxxx.joblib --mode ["svdn", "svdne"]')
- sys.exit()
- elif o in ("-i", "--image"):
- p_img_file = os.path.join(os.path.join(os.path.dirname(__file__),'../'), a)
- elif o in ("-t", "--interval"):
- p_interval = list(map(int, a.split(',')))
- elif o in ("-m", "--model"):
- p_model_file = os.path.join(os.path.join(os.path.dirname(__file__),'../'), a)
- elif o in ("-o", "--mode"):
- p_mode = a
- if p_mode != 'svdn' and p_mode != 'svdne' and p_mode != 'svd':
- assert False, "Mode not recognized"
- else:
- assert False, "unhandled option"
- # load of model file
- model = joblib.load(p_model_file)
- # load image
- img = Image.open(p_img_file)
-
- img_gray = np.array(color.rgb2gray(np.asarray(img))*255, 'uint8')
- img_mscn = image_processing.calculate_mscn_coefficients(img_gray, 7)
- img_mscn_norm = image_processing.normalize_2D_arr(img_mscn)
- img_mscn_gray = np.array(img_mscn_norm*255, 'uint8')
- SVD_MSCN = metrics.get_SVD_s(img_mscn_gray)
- # check mode to normalize data
- if p_mode == 'svdne':
-
- # need to read min_max_file
- file_path = os.path.join(os.path.join(os.path.dirname(__file__),'../'), min_max_file_path)
- with open(file_path, 'r') as f:
- min = float(f.readline().replace('\n', ''))
- max = float(f.readline().replace('\n', ''))
- l_values = image_processing.normalize_arr_with_range(SVD_MSCN, min, max)
- elif p_mode == 'svdn':
- l_values = image_processing.normalize_arr(SVD_MSCN)
- else:
- l_values = SVD_MSCN
-
- # get interval values
- begin, end = p_interval
- test_data = l_values[begin:end]
- # get prediction of model
- prediction = model.predict([test_data])[0]
- print(prediction)
- if __name__== "__main__":
- main()
|