def start():
global inputdir, outputdir
helper.initialize_sqli()
image_list = list(helper.list_files(inputdir, "image/jpeg"))
for filename in image_list:
print("Processing %s" % (filename, ))
# Creation of the SQLite row for the file
helper.image_row("evidences", filename)
extractor.basic_info(filename)
if gps.get():
extractor.PIL_exif_data_GPS(filename)
if md5.get():
extractor.md5(filename)
if sha256.get():
extractor.sha256(filename)
if sha512:
extractor.sha512(filename)
if exif.get():
extractor.exif_info(filename)
helper.create_csv(outputdir)
def generate_xml_files_with_nn(test_images_path):
"""
Bemeneti képekre futtat egy dtektálást majd ezeket egy XML fájlba menti ki.
A modellt amivel a detektálást hajtja végra azt ki lehet cserélni nagyobb modellre.
"""
# What model to download.
MODEL_NAME = 'ssd_mobilenet_v1_coco_2017_11_17'
MODEL_FILE = MODEL_NAME + '.tar.gz'
DOWNLOAD_BASE = 'http://download.tensorflow.org/models/object_detection/'
# Path to frozen detection graph. This is the actual model that is used for the object detection.
PATH_TO_CKPT = MODEL_NAME + '/frozen_inference_graph.pb'
# List of the strings that is used to add correct label for each box.
PATH_TO_LABELS = os.path.join('data', 'mscoco_label_map.pbtxt')
NUM_CLASSES = 90
if not os.path.isdir("./ssd_mobilenet_v1_coco_2017_11_17"):
opener = urllib.request.URLopener()
opener.retrieve(DOWNLOAD_BASE + MODEL_FILE, MODEL_FILE)
tar_file = tarfile.open(MODEL_FILE)
for file in tar_file.getmembers():
file_name = os.path.basename(file.name)
if 'frozen_inference_graph.pb' in file_name:
tar_file.extract(file, os.getcwd())
detection_graph = tf.Graph()
with detection_graph.as_default():
od_graph_def = tf.GraphDef()
with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:
serialized_graph = fid.read()
od_graph_def.ParseFromString(serialized_graph)
tf.import_graph_def(od_graph_def, name='')
label_map = label_map_util.load_labelmap(PATH_TO_LABELS)
categories = label_map_util.convert_label_map_to_categories(
label_map, max_num_classes=NUM_CLASSES, use_display_name=True)
category_index = label_map_util.create_category_index(categories)
# Size, in inches, of the output images.
IMAGE_SIZE = (24, 16)
i = 0
size = helper.num_of_files_in_dir(test_images_path, "jpg")
for image_path in helper.list_files(test_images_path, "jpg"):
with detection_graph.as_default():
with tf.Session() as sess:
print("Status-------------> " + str(i) + "/" + str(size))
i = i + 1
full_image_path = test_images_path + "/" + image_path
print("Opening -> ", full_image_path)
start = timeit.default_timer()
image = Image.open(full_image_path)
image_np = load_image_into_numpy_array(image)
image_np_expanded = np.expand_dims(image_np, axis=0)
all_detection_dict = run_inference_for_single_image(
sess, image_np, detection_graph)
write_objects_detected_to_xml(
get_valid_detections(all_detection_dict), category_index,
full_image_path)
stop = timeit.default_timer()
print('Detection and xml creation took: ' + str(stop - start) +
" sec")
def main(args=None):
print """\
##################################################
# imago.py #
# Digital evidences from images! #
# Made with <3 by Matteo Redaelli #
# Twitter: @solventred #
##################################################
"""
if args is None:
args = sys.argv[1:]
parser = argparse.ArgumentParser()
parser.add_argument('-i','--input', help='Input directory path', type=str, required=True)
parser.add_argument('-x','--exif', help='Extract exif metadata', action='store_true')
parser.add_argument('-g','--gps', help='Extract, parse and convert to coordinates, GPS exif metadata from images (if any)It works only with JPEG.', action='store_true')
parser.add_argument('-e','--ela', help='Extract, Error Level Analysis image,It works only with JPEG. *BETA*', action='store_true')
parser.add_argument('-n','--nude', help='Detect Nudity, It works only with JPEG, *BETA*', action='store_true')
parser.add_argument('-d','--digest', help='Calculate perceptual image hashing', type=str, choices=["md5", "sha256", "sha512", "all"])
parser.add_argument('-p','--percentualhash', help='Calculate hash digest', type=str, choices=["ahash", "phash", "dhash","whash","all"])
parser.add_argument('-o','--output', help='Output directory path', type=str)
parser.add_argument('-s','--sqli', help='Keep SQLite file after the computation', action='store_true')
parser.add_argument('-t','--type', help='Select the image, this flag can be JPEG or TIFF, if this argument it is not provided, imago will process all the image types(i.e. JPEG, TIFF)', type=str, choices=["jpeg","tiff"])
args = parser.parse_args()
if (args.exif or args.gps or args.ela or args.digest or args.nude or args.percentualhash):
filetype = ""
if (args.type == "jpeg"):
filetype = "image/jpeg"
elif (args.type == "tiff"):
filetype = "image/tiff"
else:
filetype = "image"
if args.output:
output_path = args.output
else:
output_path = "."
base_dir = args.input
helper.initialize_sqli()
image_list = list(helper.list_files(base_dir, filetype))
for filename in image_list:
print ("Processing %s" % (filename,))
# Creation of the SQLite row for the file
helper.image_row("evidences", filename)
extractor.basic_info(filename)
if args.nude:
extractor.detect_nudity(filename)
if args.gps:
extractor.PIL_exif_data_GPS(filename)
if args.percentualhash == "ahash":
extractor.ahash(filename)
elif args.percentualhash == "phash":
extractor.phash(filename)
elif args.percentualhash == "dhash":
extractor.dhash(filename)
elif args.percentualhash == "whash":
extractor.whash(filename)
elif args.percentualhash == "all":
extractor.ahash(filename)
extractor.phash(filename)
extractor.whash(filename)
extractor.dhash(filename)
if args.digest == "md5":
extractor.md5(filename)
elif args.digest == "sha256":
extractor.sha256(filename)
elif args.digest == "sha512":
extractor.sha512(filename)
elif args.digest == "all":
extractor.md5(filename)
extractor.sha256(filename)
extractor.sha512(filename)
if args.exif:
extractor.exif_info(filename)
if args.ela:
extractor.ela(filename,output_path)
print ("Processing of %s completed!" % (filename,))
# Creation of the file CSV
helper.create_csv(output_path)
if not args.sqli:
os.remove('metadata.db')
elif args.sqli and args.output:
os.rename("metadata.db", os.path.join(args.output,"metadata.db"))
else:
print("ERROR: Select at least one type of extraction")
def _get_files(img_dir):
files = helper.list_files(img_dir)
return [os.path.join(img_dir, x) for x in files]
