def compute_features_images_folder(self):
if self.model is None:
raise Exception("No model is defined")
else:
folder = os.path.join(
Path(os.path.dirname(__file__)).parent, self.folder_path)
files_in_directory = misc.get_images_in_directory(folder)
features_image_folder = []
for file, path in tqdm(files_in_directory.items()):
image_feature = self.compute_image_features(path,
print_arr=False)
features_image_folder.append(image_feature)
images = list(files_in_directory.keys())
folder_images_features_dict = {}
for i in range(len(images)):
folder_images_features_dict[
images[i]] = features_image_folder[i]
# print(folder_images_features_dict)
if self.model_name == 'SIFT':
misc.save2pickle(folder_images_features_dict,
os.path.dirname(__file__),
feature=self.model_name + "_OLD")
folder_images_features_dict_sift_new = self.compute_sift_new_features(
folder_images_features_dict)
misc.save2pickle(folder_images_features_dict_sift_new,
os.path.dirname(__file__),
feature=self.model_name)
else:
misc.save2pickle(folder_images_features_dict,
os.path.dirname(__file__),
feature=self.model_name)
def get_features(self, label):
test_dataset_folder_path = os.path.abspath(
os.path.join(Path(os.getcwd()).parent, self.labelled_dataset_path))
images_list = list(
misc.get_images_in_directory(test_dataset_folder_path).keys())
metadata = Metadata(images_list)
if self.feature_name != 'SIFT':
metadata.save_label_decomposed_features(label,
self.decomposed_feature)
features = misc.load_from_pickle(
self.reduced_pickle_file_folder,
self.decomposed_feature + '_' + label)
else:
features = {}
database_features = misc.load_from_pickle(
self.main_pickle_file_folder, self.feature_name)
label_images_list = metadata.get_specific_metadata_images_list(
feature_dict={'aspectOfHand': label})
for image in label_images_list:
features[image] = database_features[image]
return features
def get_unlabelled_images_decomposed_features(self):
test_dataset_folder_path = os.path.abspath(
os.path.join(
Path(os.getcwd()).parent, self.unlabelled_dataset_path))
images_list = list(
misc.get_images_in_directory(test_dataset_folder_path).keys())
images_decomposed_features = {}
for image_id in images_list:
features_images = FeaturesImages(self.feature_name,
test_dataset_folder_path)
test_image_path = os.path.join(test_dataset_folder_path, image_id)
test_image_features = list()
test_image_features.append(
features_images.compute_image_features(test_image_path))
if self.decomposition_name != '':
decomposed_features = self.decomposition.decomposition_model.get_new_image_features_in_latent_space(
test_image_features)
images_decomposed_features[image_id] = decomposed_features
else:
images_decomposed_features[image_id] = test_image_features
return images_decomposed_features
m = int(input("Please specify the value of m: "))
decomposition = Decomposition(decomposition_model, k, model,
test_dataset_path)
similarity = Similarity(model, image_id, m)
similarity.get_similar_images(test_dataset_path,
decomposition,
reduced_dimension=True)
elif task == '3':
model = input("1.CM\n2.LBP\n3.HOG\n4.SIFT\nSelect model: ")
decomposition_model = input(
"1.PCA\n2.SVD\n3.NMF\n4.LDA\nSelect decomposition: ")
test_dataset_folder_path = os.path.abspath(
os.path.join(Path(os.getcwd()).parent, test_dataset_path))
images_list = list(
misc.get_images_in_directory(test_dataset_folder_path).keys())
metadata = Metadata(images_list)
label = int(
input("1.Left-Hand\n2.Right-Hand\n3.Dorsal\n4.Palmar\n"
"5.With accessories\n6.Without accessories\n7.Male\n8.Female\n"
"Please choose an option: "))
label_interpret_dict = {
1: {
"aspectOfHand": "left"
},
2: {
"aspectOfHand": "right"
},
3: {
"aspectOfHand": "dorsal"
},
def get_images_list(folder_path):
folder = os.path.join(Path(os.path.dirname(__file__)).parent, folder_path)
return misc.get_images_in_directory(folder)
return class_labels
labelled_set_path = input("Enter the path for labelled input set - ")
unlabelled_set_path = input("Enter the path for unlabelled input set - ")
metadata_file_path = input("Enter the path for metadata csv - ")
metadata_file_name = input("Enter the name for labelled metadata csv - ")
unlabelled_metadata_file_name = metadata_file_name
# unlabelled_metadata_file_name = input("Enter the name for unlabelled metadata csv - ")
metadata_df = pd.read_csv(metadata_file_path+'/'+metadata_file_name)
unlabelled_metadata_df = pd.read_csv(metadata_file_path+'/'+unlabelled_metadata_file_name)
# print(metadata_df)
unlabelled_images_list = list(misc.get_images_in_directory(unlabelled_set_path).keys())
reduced_pickle_file_folder = os.path.join(os.path.dirname(os.path.abspath(__file__)), 'pickle_files')
classifier = input("Choose the Classifier - \n. 1. SVM \n. 2. Decision Tree \n. 3. PPR Classifier \n. 4. Exit \n. ")
def svm():
model = "HOG"
decomposition_model = "PCA"
phase = input("Choose from \n. 1. Train \n. 2. Test \n.")
if (phase == "train"):
# phase = "train"
training_features,images_list = test.compute_features_folder(labelled_set_path, phase)