def compile_txt_results(algorithm_info, source_db):
algorithm_uuid = algorithm_info[lookup.uuid_descriptor]
asset_dir = abspath(lookup.get_algo_asset_dirs()[lookup.detector])
image_files = lookup.get_image_list(source_db)
result_file_func = lambda file: join(asset_dir, algorithm_uuid + '_' + fs.get_filename(file[lookup.file_path], extension=False) + '.txt')
result_files = [{lookup.file_path: result_file_func(file), lookup.label: file[lookup.label]} for file in image_files]
results = []
for idx, result_file_info in enumerate(result_files):
file_result = fs.read_txt_file(result_file_info[lookup.file_path])
file_result = ''.join(file_result)
result = None
if algorithm_info[lookup.DETECTOR_TYPE] == lookup.binary_detector:
yes_filter = algorithm_info[lookup.regex_filter_yes]
no_filter = algorithm_info[lookup.regex_filter_no]
stego = re.search(yes_filter, file_result)
cover = re.search(no_filter, file_result)
assert (stego or cover and not (stego and cover))
if stego:
result = lookup.stego
else:
result = lookup.cover
else:
result = float(file_result)
result_file_info.update({lookup.result: result})
results.append(result_file_info)
return results
def load_data_as_array(db_uuid):
db_information = lookup.get_db_info(db_uuid)
image_dict = lookup.get_image_list(db_information[lookup.uuid_descriptor])
image_info = [(img.get_image_array(img_dict[lookup.file_path]),
translate_label(img_dict[lookup.label]))
for img_dict in image_dict]
return image_info
def embed_ratio(self, db_name:str, source_db:str, embedding_ratio:float):
"""generates a test DB. if divided, embeddors are randomly distributed each of the db images. otherwise each image undergoes an operation by each embeddor"""
db_information = lookup.get_source_db_info(source_db)
db_compatible_states = set(db_information[lookup.compatible_descriptor])
db_embed_compatible = db_compatible_states.intersection(self.compatible_types)
if len(db_embed_compatible) <= 0:
raise ValueError('The embeddor set and dataset are not compatible')
if embedding_ratio > self.max_embedding_ratio:
raise ValueError('The embeddor set cannot support this embedding ratio')
image_dict = lookup.get_image_list(db_information[lookup.uuid_descriptor])
image_dict = list(filter(lambda img_info: img_info[lookup.image_type] in db_embed_compatible, image_dict))
random.shuffle(image_dict)
num_images = len(image_dict)
num_embeddors = len(self.embeddors)
input_partition = []
output_partition = []
output_directory_name = fs.get_uuid()
output_directory = abspath(join(lookup.get_db_dirs()[lookup.dataset], output_directory_name))
assert(not fs.dir_exists(output_directory))
fs.make_dir(output_directory)
images_per_embeddor = int(num_images / num_embeddors)
remainder = num_images - images_per_embeddor*num_embeddors
for i in range(num_embeddors):
start_idx = i*images_per_embeddor
end_idx = (i+1)*images_per_embeddor
input_list = image_dict[start_idx:end_idx].copy()
input_partition.append(input_list)
for idx in range(remainder):
input_partition[idx].append(image_dict[idx + num_embeddors*images_per_embeddor].copy())
ratio_embeddor = partial(generator.secret_message_from_embedding, embedding_ratio)
secret_message = [list(map(ratio_embeddor, input_list)) for input_list in input_partition]
output_partition = [generator.generate_output_list(embeddor, output_directory, input_partition[idx]) for idx, embeddor in enumerate(self.embeddors)]
partition = [[{
lookup.INPUT_IMAGE_PATH: input_partition[i][j][lookup.file_path],
lookup.OUTPUT_IMAGE_PATH: output_partition[i][j],
lookup.PAYLOAD: embedding_ratio,
lookup.SECRET_TXT_PLAINTEXT: secret_message[i][j],
lookup.PASSWORD: generator.generate_password(),
}
for j in range(len(input_partition[i]))] for i in range(num_embeddors)]
db_uuid = self.embed_db(db_name, partition, source_db, embedding_ratio)
return db_uuid
def compile_csv_results(algorithm_info, source_db): algorithm_uuid = algorithm_info[lookup.uuid_descriptor] asset_dir = abspath(lookup.get_algo_asset_dirs()[lookup.detector]) image_files = lookup.get_image_list(source_db) image_filepath = [abspath(file[lookup.file_path]) for file in image_files] result_files = [algorithm_uuid + '_' + fs.get_filename(file[lookup.file_path], extension=False) + '.csv' for file in image_files] result_files = [join(asset_dir, result_file) for result_file in result_files] raise NotImplementedError
def compile_csv_directory(algorithm_info, source_db):
algorithm_uuid = algorithm_info[lookup.uuid_descriptor]
asset_dir = abspath(lookup.get_algo_asset_dirs()[lookup.detector])
image_list = lookup.get_image_list(source_db)
directory = list(set([fs.get_directory(image[lookup.file_path]) for image in image_list]))
assert(len(directory) == 1)
directory = directory[0]
result_csv_file = algorithm_uuid + '_' + fs.get_filename(directory) + '.csv'
result_csv_file = join(asset_dir, result_csv_file)
data = fs.read_csv_file(result_csv_file)
results = []
def get_image_info(file_name):
transform = lambda img: img[lookup.file_path]
if lookup.OUTPUT_FILE in algorithm_info:
if algorithm_info[lookup.OUTPUT_FILE] == lookup.INPUT_IMAGE_NAME:
transform = lambda img: fs.get_filename(img[lookup.file_path])
filtered_list = list(filter(lambda img: transform(img) == file_name, image_list))
assert(len(filtered_list) == 1)
return filtered_list[0]
for result in data:
result_info = get_image_info(result[0])
file_result = result[1]
if algorithm_info[lookup.DETECTOR_TYPE] == lookup.binary_detector:
yes_filter = algorithm_info[lookup.regex_filter_yes]
no_filter = algorithm_info[lookup.regex_filter_no]
stego = re.search(yes_filter, file_result)
cover = re.search(no_filter, file_result)
assert (stego or cover and not (stego and cover))
if stego:
result = lookup.stego
else:
result = lookup.cover
else:
result = float(file_result)
result_info.update({lookup.result: result})
results.append(result_info)
return results
def verify(self, db:str):
db_info = lookup.get_steganographic_db_info(db)
db_images = lookup.get_image_list(db_info[lookup.uuid_descriptor])
#need to group properly....
#need to read parameter assets #group all config files and batch read all the information properly...
sorted_by_embeddor = defaultdict(list)
for image_info in db_images:
image_info[lookup.INPUT_IMAGE_PATH] = image_info[lookup.file_path]
sorted_by_embeddor[image_info[lookup.uuid_descriptor]].append(image_info)
all_pre_cmds = []
all_cmds = []
all_post_cmds = []
all_termination_cmds = []
all_embeddors = {}
for embeddor_uuid in sorted_by_embeddor:
embeddor = algo.get_algorithm_info(lookup.embeddor, embeddor_uuid)
all_embeddors[embeddor_uuid] = embeddor
verify_params = copy.deepcopy(sorted_by_embeddor[embeddor_uuid])
pre_cmds, cmds, post_cmds, termination_cmds = verify_cmds.generate_commands(embeddor, verify_params)
all_pre_cmds += pre_cmds
all_cmds += cmds
all_post_cmds += post_cmds
all_termination_cmds += termination_cmds
print('running pre commands')
runner.run_pool(all_pre_cmds, self.cores)
print('completed.')
print('running commands')
runner.run_pool(all_cmds, self.cores)
print('completed.')
print('running post commands.')
runner.run_pool(all_post_cmds, self.cores)
verification_results = algo_processor.verify_embedding(db, all_embeddors)
print('completed.')
print('terminating processes...')
runner.run_pool(all_termination_cmds, self.cores)
print('completed.')
return verification_results
def detect(self, test_dbs:list):
print('collecting results from following databases: ' + str(test_dbs))
results = []
for db in test_dbs:
db_info = lookup.get_db_info(db)
db_compatible = set(db_info[lookup.compatible_descriptor])
db_detect_compatible = db_compatible.intersection(self.compatible_types)
if len(db_detect_compatible) != len(db_compatible):
raise ValueError('The detector set and dataset are not compatible')
db_image_dict = lookup.get_image_list(db)
db_image_list = list(map(lambda img: {lookup.INPUT_IMAGE_PATH: img[lookup.file_path]}, db_image_dict))
db_results = self.detect_list(db_image_list, db)
results.append(db_results)
statistics = algo.calculate_statistics(*results)
return statistics
def verify_embedding(verify_db, embeddors):
embeddor_results = defaultdict(list)
image_files = lookup.get_image_list(verify_db)
for image_file in image_files:
image_file[lookup.INPUT_IMAGE_PATH] = image_file[lookup.file_path]
embeddor_uuid = image_file[lookup.uuid_descriptor]
verify_txt_file = generator.generate_verify_file(embeddors[embeddor_uuid], image_file)
asset_file_name = fs.get_filename(verify_txt_file)
asset_directory = lookup.get_algo_asset_dirs()[lookup.embeddor]
verify_file_path = abspath(join(asset_directory, asset_file_name))
data = fs.read_txt_file(verify_file_path)
if (len(data[0])) == int(image_file[lookup.secret_txt_length]):
verification_result = True
else:
verification_result = False
embeddor_results[embeddor_uuid].append({lookup.INPUT_IMAGE_PATH: image_file[lookup.INPUT_IMAGE_PATH], lookup.result: verification_result})
return embeddor_results
def generate_labels(database_uuids, output_csv_file, relative=False):
"""generates labels.csv file for a set of databases"""
db_image_list = [('cover', 'steganographic')]
label_file_directory = abspath(lookup.get_top_level_dirs()[lookup.db])
path_to_label_file = join(label_file_directory, output_csv_file)
for db in database_uuids:
db_image_dict = lookup.get_image_list(db)
if relative:
db_image_list = db_image_list + list(
map(
lambda img:
(relpath(img[lookup.source_image], label_file_directory),
relpath(img[lookup.file_path], label_file_directory)),
db_image_dict))
else:
db_image_list = db_image_list + list(
map(
lambda img:
(img[lookup.source_image], img[lookup.file_path]),
db_image_dict))
fs.write_to_csv_file(path_to_label_file, db_image_list)
print('The labels file can be found here: ' + path_to_label_file)