for model in g.config['detection_sequence']:
# instaniate the right model
# after instantiation run all files with it,
# so no need to do 2x instantiations
t_start = datetime.datetime.now()
if model == 'object':
if g.config['ml_gateway']:
m = ObjectRemote()
else:
# print ("G LOGGER {}".format(g.logger))
m = object_detection.Object(logger=g.logger, options=g.config)
elif model == 'hog':
m = hog.Hog(options=g.config)
elif model == 'face':
if g.config['ml_gateway']:
m = FaceRemote()
else:
try:
import pyzm.ml.face as face
except ImportError:
g.logger.Error(
'Error importing face recognition. Make sure you did sudo -H pip3 install face_recognition'
)
raise
m = face.Face(logger=g.logger, options=g.config, upsample_times=g.config['face_upsample_times'],
num_jitters=g.config['face_num_jitters'],
model=g.config['face_model'])
def main_handler():
# set up logging to syslog
# construct the argument parse and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument('-c', '--config', help='config file with path')
ap.add_argument('-e', '--eventid', help='event ID to retrieve')
ap.add_argument('-p',
'--eventpath',
help='path to store object image file',
default='')
ap.add_argument('-m', '--monitorid', help='monitor id - needed for mask')
ap.add_argument('-v',
'--version',
help='print version and quit',
action='store_true')
ap.add_argument(
'-o',
'--output-path',
help='internal testing use only - path for debug images to be written')
ap.add_argument('-f',
'--file',
help='internal testing use only - skips event download')
ap.add_argument('-r',
'--reason',
help='reason for event (notes field in ZM)')
ap.add_argument('-n',
'--notes',
help='updates notes field in ZM with detections',
action='store_true')
ap.add_argument('-d',
'--debug',
help='enables debug on console',
action='store_true')
args, u = ap.parse_known_args()
args = vars(args)
if args.get('version'):
print('hooks:{} pyzm:{}'.format(hooks_version, pyzm_version))
exit(0)
if not args.get('config'):
print('--config required')
exit(1)
if not args.get('file') and not args.get('eventid'):
print('--eventid required')
exit(1)
utils.get_pyzm_config(args)
if args.get('debug'):
g.config['pyzm_overrides']['dump_console'] = True
if args.get('monitorid'):
log.init(name='zmesdetect_' + 'm' + args.get('monitorid'),
override=g.config['pyzm_overrides'])
else:
log.init(name='zmesdetect', override=g.config['pyzm_overrides'])
g.logger = log
es_version = '(?)'
try:
es_version = subprocess.check_output(
['/usr/bin/zmeventnotification.pl', '--version']).decode('ascii')
except:
pass
try:
import cv2
except ImportError as e:
g.logger.Fatal(
f'{e}: You might not have installed OpenCV as per install instructions. Remember, it is NOT automatically installed'
)
g.logger.Info(
'---------| pyzm version: {}, ES version: {} , OpenCV version: {}|------------'
.format(__version__, es_version, cv2.__version__))
# load modules that depend on cv2
try:
import zmes_hook_helpers.image_manip as img
import pyzm.ml.alpr as alpr
except Exception as e:
g.logger.Error(f'{e}')
exit(1)
g.polygons = []
# process config file
g.ctx = ssl.create_default_context()
utils.process_config(args, g.ctx)
# misc came later, so lets be safe
if not os.path.exists(g.config['base_data_path'] + '/misc/'):
try:
os.makedirs(g.config['base_data_path'] + '/misc/')
except FileExistsError:
pass # if two detects run together with a race here
if not g.config['ml_gateway']:
g.logger.Info('Importing local classes for Object/Face')
import pyzm.ml.object as object_detection
import pyzm.ml.hog as hog
else:
g.logger.Info('Importing remote shim classes for Object/Face')
from zmes_hook_helpers.apigw import ObjectRemote, FaceRemote, AlprRemote
# now download image(s)
if not args.get('file'):
try:
filename1, filename2, filename1_bbox, filename2_bbox = utils.download_files(
args)
except Exception as e:
g.logger.Error(f'Error downloading files: {e}')
g.logger.Fatal('error: Traceback:{}'.format(
traceback.format_exc()))
# filename_alarm will be the first frame to analyze (typically alarm)
# filename_snapshot will be the second frame to analyze only if the first fails (typically snapshot)
else:
g.logger.Debug(
1, 'TESTING ONLY: reading image from {}'.format(args.get('file')))
filename1 = args.get('file')
filename1_bbox = g.config['image_path'] + '/' + append_suffix(
filename1, '-bbox')
filename2 = None
filename2_bbox = None
start = datetime.datetime.now()
obj_json = []
# Read images to analyze
image2 = None
image1 = cv2.imread(filename1)
if image1 is None: # can't have this None, something went wrong
g.logger.Error(
'Error reading {}. It either does not exist or is invalid'.format(
filename1))
raise ValueError(
'Error reading file {}. It either does not exist or is invalid'.
format(filename1))
oldh, oldw = image1.shape[:2]
if filename2: # may be none
image2 = cv2.imread(filename2)
if image2 is None:
g.logger.Error(
'Error reading {}. It either does not exist or is invalid'.
format(filename2))
raise ValueError(
'Error reading file {}. It either does not exist or is invalid'
.format(filename2))
# create a scaled polygon for object intersection checks
if not g.polygons and g.config['only_triggered_zm_zones'] == 'no':
g.polygons.append({
'name': 'full_image',
'value': [(0, 0), (oldw, 0), (oldw, oldh), (0, oldh)],
'pattern': g.config.get('object_detection_pattern')
})
g.logger.Debug(
1, 'No polygon area specfied, so adding a full image polygon:{}'.
format(g.polygons))
if g.config['resize'] != 'no':
g.logger.Debug(
1, 'resizing to {} before analysis...'.format(g.config['resize']))
image1 = imutils.resize(image1,
width=min(int(g.config['resize']),
image1.shape[1]))
if image2 is not None:
image2 = imutils.resize(image2,
width=min(int(g.config['resize']),
image2.shape[1]))
newh, neww = image1.shape[:2]
utils.rescale_polygons(neww / oldw, newh / oldh)
# Apply all configured models to each file
matched_file = None
bbox = []
label = []
conf = []
classes = []
use_alpr = True if 'alpr' in g.config['detection_sequence'] else False
g.logger.Debug(1, 'User ALPR if vehicle found: {}'.format(use_alpr))
# labels that could have license plates. See https://github.com/pjreddie/darknet/blob/master/data/coco.names
for model in g.config['detection_sequence']:
# instaniate the right model
# after instantiation run all files with it,
# so no need to do 2x instantiations
t_start = datetime.datetime.now()
if model == 'object':
if g.config['ml_gateway']:
m = ObjectRemote()
else:
# print ("G LOGGER {}".format(g.logger))
m = object_detection.Object(logger=g.logger, options=g.config)
elif model == 'hog':
m = hog.Hog(options=g.config)
elif model == 'face':
if g.config['ml_gateway']:
m = FaceRemote()
else:
try:
import pyzm.ml.face as face
except ImportError:
g.logger.Error(
'Error importing face recognition. Make sure you did sudo -H pip3 install face_recognition'
)
raise
m = face.Face(logger=g.logger,
options=g.config,
upsample_times=g.config['face_upsample_times'],
num_jitters=g.config['face_num_jitters'],
model=g.config['face_model'])
elif model == 'alpr':
if g.config['alpr_use_after_detection_only'] == 'yes':
#g.logger.Debug (1,'Skipping ALPR as it is configured to only be used after object detection')
continue # we would have handled it after object
else:
g.logger.Info(
'Standalone ALPR is not supported today. Please use after object'
)
continue
else:
g.logger.Error('Invalid model {}'.format(model))
raise ValueError('Invalid model {}'.format(model))
#g.logger.Debug(1,'|--> model:{} init took: {}s'.format(model, (datetime.datetime.now() - t_start).total_seconds()))
# read the detection pattern we need to apply as a filter
pat = model + '_detection_pattern'
try:
g.logger.Debug(
2, 'using g.config[\'{}\']={}'.format(pat, g.config[pat]))
r = re.compile(g.config[pat])
except re.error:
g.logger.Error('invalid pattern {} in {}, using .*'.format(
pat, g.config[pat]))
r = re.compile('.*')
t_start = datetime.datetime.now()
try_next_image = False # take the best of both images, currently used only by alpr
# temporary holders, incase alpr is used but not found
saved_bbox = []
saved_labels = []
saved_conf = []
saved_classes = []
saved_image = None
saved_file = None
# Apply the model to all files
remote_failed = False
# default order is alarm, snapshot
frame_order = [
filename2, filename1
] if g.config['bestmatch_order'] == 's,a' else [filename1, filename2]
for filename in frame_order:
if filename is None:
continue
#filename = './car.jpg'
if matched_file and filename != matched_file:
# this will only happen if we tried model A, we found a match
# and then we looped to model B to find more matches (that is, detection_mode is all)
# in this case, we only want to match more models to the file we found a first match
g.logger.Debug(
1, 'Skipping {} as we earlier matched {}'.format(
filename, matched_file))
continue
g.logger.Debug(1,
'Using model: {} with {}'.format(model, filename))
image = image1 if filename == filename1 else image2
original_image = image.copy()
if g.config['ml_gateway'] and not remote_failed:
try:
b, l, c = remote_detect(original_image, model)
except Exception as e:
g.logger.Error('Error executing remote API: {}'.format(e))
if g.config['ml_fallback_local'] == 'yes':
g.logger.Info('Falling back to local execution...')
remote_failed = True
if model == 'object':
import pyzm.ml.object as object_detection
m = object_detection.Object(logger=g.logger,
options=g.config)
elif model == 'hog':
import pyzm.ml.hog as hog
m = hog.Hog(options=g.config)
elif model == 'face':
import pyzm.ml.face as face
m = face.Face(
options=g.config,
upsample_times=g.config['face_upsample_times'],
num_jitters=g.config['face_num_jitters'],
model=g.config['face_model'])
b, l, c = m.detect(original_image)
else:
raise
else:
b, l, c = m.detect(original_image)
#g.logger.Debug(1,'|--> model:{} detection took: {}s'.format(model,(datetime.datetime.now() - t_start).total_seconds()))
t_start = datetime.datetime.now()
# Now look for matched patterns in bounding boxes
match = list(filter(r.match, l))
# If you want face recognition, we need to add the list of found faces
# to the allowed list or they will be thrown away during the intersection
# check
if model == 'face':
match = match + [g.config['unknown_face_name']] # unknown face
if g.config['ml_gateway'] and not remote_failed:
data_file = g.config[
'base_data_path'] + '/misc/known_face_names.json'
if os.path.exists(data_file):
g.logger.Debug(
1,
'Found known faces list remote gateway supports. If you have trained new faces in the remote gateway, please delete this file'
)
with open(data_file) as json_file:
data = json.load(json_file)
g.logger.Debug(
2, 'Read from existing names: {}'.format(
data['names']))
m.set_classes(data['names'])
else:
g.logger.Debug(
1, 'Fetching known names from remote gateway')
api_url = g.config[
'ml_gateway'] + '/detect/object?type=face_names'
r = requests.post(url=api_url,
headers=auth_header,
params={})
data = r.json()
with open(data_file, 'w') as json_file:
wdata = {'names': data['names']}
json.dump(wdata, json_file)
'''
for cls in m.get_classes():
if not cls in match:
match = match + [cls]
'''
# now filter these with polygon areas
#g.logger.Debug (1,"INTERIM BOX = {} {}".format(b,l))
b, l, c = img.processFilters(b, l, c, match, model)
if use_alpr:
vehicle_labels = ['car', 'motorbike', 'bus', 'truck', 'boat']
if not set(l).isdisjoint(vehicle_labels) or try_next_image:
# if this is true, that ,means l has vehicle labels
# this happens after match, so no need to add license plates to filter
g.logger.Debug(
1,
'Invoking ALPR as detected object is a vehicle or, we are trying hard to look for plates...'
)
if g.config['ml_gateway']:
alpr_obj = AlprRemote()
else:
alpr_obj = alpr.Alpr(logger=g.logger, options=g.config)
if g.config['ml_gateway'] and not remote_failed:
try:
alpr_b, alpr_l, alpr_c = remote_detect(
original_image, 'alpr')
except Exception as e:
g.logger.Error(
'Error executing remote API: {}'.format(e))
if g.config['ml_fallback_local'] == 'yes':
g.logger.Info(
'Falling back to local execution...')
remote_failed = True
alpr_obj = alpr.Alpr(logger=g.logger,
options=g.config)
alpr_b, alpr_l, alpr_c = alpr_obj.detect(
original_image)
else:
raise
else: # not ml_gateway
alpr_b, alpr_l, alpr_c = alpr_obj.detect(
original_image)
alpr_b, alpr_l, alpr_c = img.getValidPlateDetections(
alpr_b, alpr_l, alpr_c)
if len(alpr_l):
#g.logger.Debug (1,'ALPR returned: {}, {}, {}'.format(alpr_b, alpr_l, alpr_c))
try_next_image = False
# First get non plate objects
for idx, t_l in enumerate(l):
otype = 'face' if model == 'face' else 'object'
obj_json.append({
'type':
otype,
'label':
t_l,
'box':
b[idx],
'confidence':
"{:.2f}%".format(c[idx] * 100)
})
# Now add plate objects
for i, al in enumerate(alpr_l):
g.logger.Debug(
2, 'ALPR Found {} at {} with score:{}'.format(
al, alpr_b[i], alpr_c[i]))
b.append(alpr_b[i])
l.append(al)
c.append(alpr_c[i])
obj_json.append({
'type':
'licenseplate',
'label':
al,
'box':
alpr_b[i],
#'confidence': alpr_c[i]
'confidence':
"{:.2f}%".format(alpr_c[i] * 100)
})
elif filename == filename1 and filename2: # no plates, but another image to try
g.logger.Debug(
1,
'We did not find license plates in vehicles, but there is another image to try'
)
saved_bbox = b
saved_labels = l
saved_conf = c
saved_classes = m.get_classes()
saved_image = image.copy()
saved_file = filename
try_next_image = True
else: # no plates, no more to try
g.logger.Info(
'We did not find license plates, and there are no more images to try'
)
if saved_bbox:
g.logger.Debug(
2, 'Going back to matches in first image')
b = saved_bbox
l = saved_labels
c = saved_conf
image = saved_image
filename = saved_file
# store non plate objects
otype = 'face' if model == 'face' else 'object'
for idx, t_l in enumerate(l):
obj_json.append({
'type':
otype,
'label':
t_l,
'box':
b[idx],
'confidence':
"{:.2f}%".format(c[idx] * 100)
})
try_next_image = False
else: # objects, no vehicles
if filename == filename1 and filename2:
g.logger.Debug(
1,
'There was no vehicle detected by object detection in this image'
)
'''
# For now, don't force ALPR in the next (snapshot image)
# only do it if object_detection gets a vehicle there
# may change this later
try_next_image = True
saved_bbox = b
saved_labels = l
saved_conf = c
saved_classes = m.get_classes()
saved_image = image.copy()
saved_file = filename
'''
else:
g.logger.Debug(
1,
'No vehicle detected, and no more images to try')
if saved_bbox:
g.logger.Debug(
1, 'Going back to matches in first image')
b = saved_bbox
l = saved_labels
c = saved_conf
image = saved_image
filename = saved_file
try_next_image = False
otype = 'face' if model == 'face' else 'object'
for idx, t_l in enumerate(l):
obj_json.append({
'type':
'object',
'label':
t_l,
'box':
b[idx],
'confidence':
"{:.2f}%".format(c[idx] * 100)
})
else: # usealpr
g.logger.Debug(
2,
'ALPR not in use, no need for look aheads in processing')
# store objects
otype = 'face' if model == 'face' else 'object'
for idx, t_l in enumerate(l):
obj_json.append({
'type':
otype,
'label':
t_l,
'box':
b[idx],
'confidence':
"{:.2f}%".format(c[idx] * 100)
})
if b:
# g.logger.Debug (1,'ADDING {} and {}'.format(b,l))
if not try_next_image:
bbox.extend(b)
label.extend(l)
conf.extend(c)
classes.append(m.get_classes())
g.logger.Info('labels found: {}'.format(l))
g.logger.Debug(
2, 'match found in {}, breaking file loop...'.format(
filename))
matched_file = filename
break # if we found a match, no need to process the next file
else:
g.logger.Debug(
2,
'Going to try next image before we decide the best one to use'
)
else:
g.logger.Debug(
1, 'No match found in {} using model:{}'.format(
filename, model))
# file loop
# model loop
if matched_file and g.config['detection_mode'] == 'first':
g.logger.Debug(
2,
'detection mode is set to first, breaking out of model loop...'
)
break
# all models loops, all files looped
#g.logger.Debug (1,'FINAL LIST={} AND {}'.format(bbox,label))
# Now create prediction string
pred = ''
if not matched_file:
g.logger.Info('No patterns found using any models in all files')
else:
# we have matches
if matched_file == filename1:
#image = image1
bbox_f = filename1_bbox
else:
#image = image2
bbox_f = filename2_bbox
# let's remove past detections first, if enabled
if g.config['match_past_detections'] == 'yes' and args.get(
'monitorid'):
# point detections to post processed data set
g.logger.Info('Removing matches to past detections')
bbox_t, label_t, conf_t = img.processPastDetection(
bbox, label, conf, args.get('monitorid'))
# save current objects for future comparisons
g.logger.Debug(
1, 'Saving detections for monitor {} for future match'.format(
args.get('monitorid')))
try:
mon_file = g.config['image_path'] + '/monitor-' + args.get(
'monitorid') + '-data.pkl'
f = open(mon_file, "wb")
pickle.dump(bbox, f)
pickle.dump(label, f)
pickle.dump(conf, f)
f.close()
except Exception as e:
g.logger.Error(
f'Error writing to {mon_file}, past detections not recorded:{e}'
)
bbox = bbox_t
label = label_t
conf = conf_t
# now we draw boxes
g.logger.Debug(2, "Drawing boxes around objects")
out = img.draw_bbox(image, bbox, label, classes, conf, None,
g.config['show_percent'] == 'yes')
image = out
if g.config['frame_id'] == 'bestmatch':
if matched_file == filename1:
prefix = '[a] ' # we will first analyze alarm
frame_type = 'alarm'
else:
prefix = '[s] '
frame_type = 'snapshot'
else:
prefix = '[x] '
frame_type = g.config['frame_id']
if g.config['write_debug_image'] == 'yes':
g.logger.Debug(
1,
'Writing out debug bounding box image to {}...'.format(bbox_f))
cv2.imwrite(bbox_f, image)
# Do this after match past detections so we don't create an objdetect if images were discarded
if g.config['write_image_to_zm'] == 'yes':
if (args.get('eventpath') and len(bbox)):
g.logger.Debug(
1, 'Writing detected image to {}/objdetect.jpg'.format(
args.get('eventpath')))
cv2.imwrite(args.get('eventpath') + '/objdetect.jpg', image)
jf = args.get('eventpath') + '/objects.json'
final_json = {'frame': frame_type, 'detections': obj_json}
g.logger.Debug(1, 'Writing JSON output to {}'.format(jf))
try:
with open(jf, 'w') as jo:
json.dump(final_json, jo)
jo.close()
except Exception as e:
g.logger.Error(f'Error creating {jf}:{e}')
if g.config['create_animation'] == 'yes':
g.logger.Debug(1, 'animation: Creating burst...')
try:
img.createAnimation(
frame_type, args.get('eventid'),
args.get('eventpath') + '/objdetect',
g.config['animation_types'])
except Exception as e:
g.logger.Error('Error creating animation:{}'.format(e))
g.logger.Error('animation: Traceback:{}'.format(
traceback.format_exc()))
else:
if not len(bbox):
g.logger.Debug(
1, 'Not writing image, as no objects recorded')
else:
g.logger.Error(
'Could not write image to ZoneMinder as eventpath not present'
)
detections = []
seen = {}
if not obj_json:
# if we broke out early/first match
otype = 'face' if model == 'face' else 'object'
for idx, t_l in enumerate(label):
#print (idx, t_l)
obj_json.append({
'type': otype,
'label': t_l,
'box': bbox[idx],
'confidence': "{:.2f}%".format(conf[idx] * 100)
})
#g.logger.Debug (1,'CONFIDENCE ARRAY:{}'.format(conf))
for idx, l in enumerate(label):
if l not in seen:
if g.config['show_percent'] == 'no':
pred = pred + l + ','
else:
pred = pred + l + ':{:.0%}'.format(conf[idx]) + ' '
seen[l] = 1
if pred != '':
pred = pred.rstrip(',')
pred = prefix + 'detected:' + pred
g.logger.Info('Prediction string:{}'.format(pred))
# g.logger.Error (f"Returning THIS IS {obj_json}")
jos = json.dumps(obj_json)
g.logger.Debug(1, 'Prediction string JSON:{}'.format(jos))
print(pred + '--SPLIT--' + jos)
# end of matched_file
if g.config['delete_after_analyze'] == 'yes':
try:
if filename1:
os.remove(filename1)
if filename2:
os.remove(filename2)
except Exception as e:
g.logger.Error(f'Could not delete file(s):{e}')
if args.get('notes') and pred:
# We want to update our DB notes with the detection string
g.logger.Debug(1,
'Updating notes for EID:{}'.format(args.get('eventid')))
import pyzm.api as zmapi
api_options = {
'apiurl': g.config['api_portal'],
'portalurl': g.config['portal'],
'user': g.config['user'],
'password': g.config['password'],
'logger': g.logger # We connect the API to zmlog
#'logger': None, # use none if you don't want to log to ZM,
#'disable_ssl_cert_check': True
}
try:
myapi = zmapi.ZMApi(options=api_options)
except Exception as e:
g.logger.Error('Error during login: {}'.format(str(e)))
g.logger.Debug(2, traceback.format_exc())
exit(0) # Let's continue with zmdetect
url = '{}/events/{}.json'.format(g.config['api_portal'],
args['eventid'])
try:
ev = myapi._make_request(url=url, type='get')
except Exception as e:
g.logger.Error('Error during event notes retrieval: {}'.format(
str(e)))
g.logger.Debug(2, traceback.format_exc())
exit(0) # Let's continue with zmdetect
new_notes = pred
if ev.get('event', {}).get('Event', {}).get('Notes'):
old_notes = ev['event']['Event']['Notes']
old_notes_split = old_notes.split('Motion:')
old_d = old_notes_split[0] # old detection
try:
old_m = old_notes_split[1]
except IndexError:
old_m = ''
new_notes = pred + 'Motion:' + old_m
g.logger.Debug(
1, 'Replacing old note:{} with new note:{}'.format(
old_notes, new_notes))
payload = {}
payload['Event[Notes]'] = new_notes
try:
ev = myapi._make_request(url=url, payload=payload, type='put')
except Exception as e:
g.logger.Error('Error during notes update: {}'.format(str(e)))
g.logger.Debug(2, traceback.format_exc())
exit(0) # Let's continue with zmdetect