def video_img():
detector = Detector()
vc=cv2.VideoCapture(r'F:\video\mv.mp4')
c=1
if vc.isOpened():
rval,frame=vc.read()
else:
rval=False
while rval:
rval,frame=vc.read()
im = Image.fromarray(cv2.cvtColor(frame,cv2.COLOR_BGR2RGB))
boxes = detector.detect(im)
for box in boxes:
x1 = int(box[0])
y1 = int(box[1])
x2 = int(box[2])
y2 = int(box[3])
_y2 = int(y2 * 0.8+y1*0.2)
cv2.rectangle(frame,(x1, y1), (x2, _y2), (0,225,0),2)
cv2.imshow('img',frame)
# cv2.imwrite(r'F:\video\img\img'+str(c)+'.jpg',frame)
c=c+1
cv2.waitKey(1)
vc.release()
def main():
parser = ap.ArgumentParser(
description=
"Arguments to use in data collection and Arduino communication.")
parser.add_argument("port", help="The serial port file to use.")
args = parser.parse_args()
serial_port = args.port
if not os.path.exists(serial_port):
print("Error: file '{}' does not exist or is not a file.".format(
serial_port))
exit(1)
# William's test driver code, but with the serial port as a command line arg.
# ctl = ac(serial_port)
# need this delay for arduino initialization
time.sleep(3)
print('Starting detection...')
detector = Detector(serial_port)
while True:
try:
print('writing: ' + str(detector.sample()))
except Exception as e:
print(e)
print('Breaking')
detector.close()
break
def run(datafile, data_dir):
def download(url):
try:
with urlopen(url) as response:
if response.status == 200:
data = np.asarray(bytearray(response.read()),
dtype=np.uint8)
return cv2.imdecode(data, cv2.IMREAD_COLOR)
except HTTPError as e:
print(f'{url}: {e}')
except (ConnectionError, OSError) as e:
print(f'{url}: {e}')
time.sleep(0.1)
except Exception as e:
print(f'{url}: {e}')
time.sleep(0.5)
return None
detector = Detector()
with open(datafile, 'r') as fp:
r = csv.reader(fp, delimiter='\t')
for row in r:
photo_url = row[1]
print(photo_url)
img = download(photo_url)
if img is None:
continue
result = detector.detect(img)
if result is None:
continue
basename = os.path.basename(photo_url)
path0 = f'{ord(basename[0]):02x}'
path1 = f'{ord(basename[1]):02x}'
path2 = f'{ord(basename[2]):02x}'
outdir = os.path.join(data_dir, path0, path1, path2)
os.makedirs(outdir, exist_ok=True)
faceimg = result['image']
del result['image']
created_at = datetime.fromtimestamp(
time.mktime(time.strptime(row[3],
'%a %b %d %H:%M:%S +0000 %Y')))
result['meta'] = {
'photo_id': row[0],
'photo_url': row[1],
'source_url': row[2],
'published_at': created_at.isoformat(),
'label_id': row[4],
'label_name': row[5],
}
name = os.path.splitext(basename)[0]
cv2.imwrite(os.path.join(outdir, f'{name}.jpg'), faceimg,
[cv2.IMWRITE_JPEG_QUALITY, 100])
with open(os.path.join(outdir, f'{name}.json'), 'w') as fp:
json.dump(result, fp, ensure_ascii=False)
def __init__(self, opt):
super(mywindow, self).__init__()
self.setupUi(self)
self.setFixedSize(self.width(), self.height())
self.opt = opt
self.detector = Detector(cfg=opt.cfg,
data_cfg=opt.data_cfg,
weights=opt.weights,
img_size=opt.img_size,
conf_thres=opt.conf_thres,
nms_thres=opt.nms_thres)
def main():
args = parseopt()
dump_flow(args.movie)
flow, header = pick_flow(args.movie)
detector = Detector(gpu=args.gpu)
draw_box_flow_func = \
lambda movie, flow: draw_box_flow(movie, flow, detector)
if args.nobbox:
vis_flow(args.movie, flow)
else:
vis_flow(args.movie, flow, draw=draw_box_flow_func)
def main_func(path):
"""
Detect 3rd-party libraries in app.
The detection result is printed.
:param path: The path of target app.
:return: None.
"""
print "--Decoding--"
detector = Detector()
decoded_path = detector.get_smali(path)
detector.get_hash(decoded_path)
def __init__(self):
handlers = [
(r"/config", ConfigHandler),
(r"/voice", VoiceHandler),
(r"/detect", HttpDetectHandler),
]
settings = dict(
debug=options.dev,
static_path=os.path.join(
os.path.dirname(os.path.abspath(__file__)), "static"),
template_path=os.path.join(
os.path.dirname(os.path.abspath(__file__)), "templates"),
)
super(Application, self).__init__(handlers, **settings)
self.data = None
self.color_map = collections.defaultdict(lambda: (random.randint(
0, 255), random.randint(0, 255), random.randint(0, 255)))
for i in range(1, 21):
self.color_map[str(i)] = [
random.randint(0, 255),
random.randint(0, 255),
random.randint(0, 255)
]
if options.model.find('det300') >= 0:
self.model_det300 = Detector('conf/retina_net.conf')
if options.model.find('det600') >= 0:
self.model_det600 = Detector('conf/retina_net_600.conf')
if options.model.find('mask600') >= 0:
self.model_mask600 = MaskRetinaNet('conf/mask_600.conf')
if options.model.find('seg512') >= 0:
self.model_seg512 = Segmentor('conf/cloth_320.conf', rotate=False)
self.model_det300 = Detector('conf/retina_net.conf')
self.model_fat = FatDetector()
self.model_multi_task = MultiTaskDetector()
logging.info("initialize done")
def __init__(self, minutes):
queue = Queue(maxsize=100)
self.queue = queue
self.recorder = Recorder()
self.minutes = minutes
self.consoleAudioHandler = ConsoleAudioHandler(self.queue,
config["tmpPath"])
t = Thread(target=self.worker)
t.start()
provider = ConsoleAudioProvider(self.queue)
detector = Detector(provider)
detector.detect()
def run(data_dir):
detector = Detector()
with open('data.tsv', 'r') as fp:
r = csv.reader(fp, delimiter='\t')
next(r)
for row in r:
photo_url = row[3]
basename = os.path.basename(photo_url)
path0 = f'{ord(basename[0]):02x}'
path1 = f'{ord(basename[1]):02x}'
path2 = f'{ord(basename[2]):02x}'
filepath = os.path.join(data_dir, 'images', path0, path1, path2,
basename)
print(f'processing {filepath} ...')
try:
result = detector.detect(cv2.imread(filepath))
if result is None:
print('detection failed.')
continue
outdir = os.path.join(data_dir, 'results', path0, path1, path2)
os.makedirs(outdir, exist_ok=True)
img = result['image']
del result['image']
result['meta'] = {
'face_id': row[0],
'photo_id': row[1],
'source_url': row[2],
'photo_url': row[3],
'posted_at': row[4],
'label_id': row[5],
'label_name': row[6],
}
name = os.path.splitext(basename)[0]
cv2.imwrite(os.path.join(outdir, f'{name}.png'), img)
with open(os.path.join(outdir, f'{name}.json'), 'w') as fp:
json.dump(result, fp, ensure_ascii=False)
except:
print(f'error!!: {filepath}')
time.sleep(1)
def track(video_path, use_gpu=False):
video = cv2.VideoCapture(video_path)
ret, frame = video.read()
if ret:
frame = cv2.resize(frame, (input_width, input_height))
if use_gpu:
caffe.set_mode_gpu()
tracker = Sort(max_age=10)
detector = Detector()
classes = detector.get_classes()
while ret:
frame_disp = np.copy(frame)
bounding_boxes, counting = detector.infer(frame)
class_counting = zip(classes, counting)
for pair in class_counting:
print('{:s} {:03d}'.format(*pair))
print('')
if len(bounding_boxes) > 0:
bounding_boxes = np.array(bounding_boxes, np.int32)
# convert (x, y, w, h) to (x1, y1, x2, y2)
bounding_boxes[:, 2:4] += bounding_boxes[:, 0:2]
bounding_boxes[:, 2:4] -= 1
track_results = tracker.update(bounding_boxes)
draw_tracking_results(track_results, frame_disp)
cv2.imshow('tracking', frame_disp)
key = cv2.waitKey(1)
if key == 27:
return
ret, frame = video.read()
if ret:
frame = cv2.resize(frame, (input_width, input_height))
def result_video():
if request.method == 'POST':
if 'file' not in request.files:
flash('No file part')
return redirect(request.url)
file = request.files['file']
if file.filename == '':
flash('No selected file')
return redirect(request.url)
if file and allowed_file(file.filename):
filename = secure_filename(file.filename)
filename = os.path.join(UPLOAD_FOLDER, filename)
file.save(filename)
model = Detector(filename, 1)
model.run_detector()
model.generate_plots()
model.write_video()
return render_template('result_video.html')
#return redirect(url_for('uploaded_file',filename=filename))
return render_template('upload_video.html')
if loc is None:
loc = Locator(frame)
# loc.locate_tube()
# loc.locate_window()
# loc.locate_scale()
loc.tube = [(0, 271, 33, 76), (0, 271, 78, 115)]
loc.window = [[(47, 80, 43, 61), (82, 115, 42, 58),
(117, 151, 39, 58), (153, 186, 41, 61),
(188, 220, 43, 61)],
[(2, 35, 88, 108), (35, 69, 85, 105),
(72, 105, 84, 105), (108, 142, 84, 104),
(145, 178, 84, 104), (179, 213, 87, 108),
(214, 246, 89, 109)]]
loc.scale = [[110, 199], [102, 199]]
tube0 = Detector(loc.tube[0], loc.window[0], loc.scale[0], 210, 50)
tube1 = Detector(loc.tube[1], loc.window[1], loc.scale[1], 160, 33)
tube0.feed(frame)
tube1.feed(frame)
if not count % 20:
tube0.update_backgrounds()
tube1.update_backgrounds()
level = int(tube0.cur_level)
cv2.line(frame, (45, level), (60, level), (0, 0, 255), 1)
level = int(tube1.cur_level)
cv2.line(frame, (90, level), (105, level), (0, 0, 255), 1)
text0 = 'level0: ' + str(round(tube0.level_scale))
text1 = 'level1: ' + str(round(tube1.level_scale))
cv2.putText(frame, text0, (5, 13), cv2.FONT_HERSHEY_COMPLEX_SMALL, 0.6,
(0, 0, 255), 1)
parser.add_argument("--reidpath", type=str, default='reidLib', help="outputdir")
opt = parser.parse_args()
videoname = os.path.join(opt.indir,opt.videoin)
output_dir = opt.outdir
work_dir = opt.workdir
reid_dir = opt.reiddir if opt.reiddir != "" else None
make_folder(output_dir)
clean_folder(work_dir)
clean_folder(reid_dir)
sys.path.append(opt.yolopath)
from detect import Detector
detector = Detector(conf_thres=opt.detconf, yolo_path=opt.yolopath, resize_size=opt.imgsize, output_dir=work_dir)
from reid import Reid
reid = Reid(save_path=reid_dir, threshold=opt.threshold, verbose=False)
videoframe = cv2.VideoCapture(videoname)
framenr = 0
if not videoframe.isOpened():
print("Error opening video stream or file")
while videoframe.isOpened():
ret, img = videoframe.read()
if not ret: # reached end of video
break
def __init__(self):
# Create the olympe.Drone object from its IP address
self.drone = olympe.Drone(DRONE_IP)
# subscribe to flight listener
listener = FlightListener(self.drone)
listener.subscribe()
self.last_frame = np.zeros((1, 1, 3), np.uint8)
self.frame_queue = queue.Queue()
self.flush_queue_lock = threading.Lock()
self.detector = Detector()
self.keypoints_image = np.zeros((1, 1, 3), np.uint8)
self.keypoints = deque(maxlen=5)
self.faces = deque(maxlen=10)
self.f = open("distances.csv", "w")
self.face_distances = deque(maxlen=10)
self.image_width = 1280
self.image_height = 720
self.half_face_detection_size = 150
self.poses_model = load("models/posesmodel.joblib")
self.pose_predictions = deque(maxlen=5)
self.pause_finding_condition = threading.Condition(threading.Lock())
self.pause_finding_condition.acquire()
self.pause_finding = True
self.person_thread = threading.Thread(target=self.fly_to_person)
self.person_thread.start()
# flight parameters in meter
self.flight_height = 0.0
self.max_height = 1.0
self.min_dist = 1.5
# keypoint map
self.nose = 0
self.left_eye = 1
self.right_eye = 2
self.left_ear = 3
self.right_ear = 4
self.left_shoulder = 5
self.right_shoulder = 6
self.left_elbow = 7
self.right_elbow = 8
self.left_wrist = 9
self.right_wrist = 10
self.left_hip = 11
self.right_hip = 12
self.left_knee = 13
self.right_knee = 14
self.left_ankle = 15
self.right_ankle = 16
# person distance
self.eye_dist = 0.0
# save images
self.save_image = False
self.image_counter = 243
self.pose_file = open("poses.csv", "w")
super().__init__()
super().start()
def __init__(self):
self.detector = Detector()
import logging
import yaml
from detect import Detector
config_dir = "./"
# Load configuration.
config_file_path = config_dir + 'config.yml'
config = yaml.safe_load(open(config_file_path, 'r'))
# start_detect(config)
if __name__ == "__main__":
print(
"====================================FCD Started===================================="
)
LOG_FORMAT = "%(asctime)s - %(levelname)s - %(message)s"
logging.basicConfig(filename='logs.log',
level=logging.DEBUG,
format=LOG_FORMAT)
detector = Detector(config)
detector.start_detect()
def __init__(self, **kwargs):
super(CalcGridLayout, self).__init__(**kwargs)
Window.bind(on_dropfile=self._on_file_drop)
self.detector = Detector(self.weights, self.filePath)
self.ids.model_label.text = self.weights_description[0]
def work(self, im):
super(HandCropper, self).work(im)
self.thresholdGrey(150)
self.findContours()
self.filterSudoku() # minAreaRect旋转矩形, 筛选出九个九宫格块的轮廓
self.alignPlank() # 透视变换, 对齐并切割出大板（包含整个九宫格和七段管数字板）
self.splitSudokuLight()
self.threshold_im = self.threshold_sudoku_im
self.findRecContours()
self.filterRecContoursByArea(9) # boundingRect, 筛选出九宫格块
self.rec_contours = self.resumeOrder(self.rec_contours[:9],
ykey=lambda it: it[1],
xkey=lambda it: it[0])
self.cropRecContours(self.rec_contours,
base_im=self.threshold_sudoku_im)
if SAVE:
self.saveCropResults()
self.saveCropResults(results=[self.threshold_light_im])
return self.results
if __name__ == '__main__':
# d = Detector(LightCropper(), None)
# d.work('../test_im/52648.jpg')
d = Detector(HandCropper(), None)
d.work('../test_im/real1.jpg')
from detect import Detector
from imgUtils import ImgUtils
if __name__ == "__main__":
param = {
(1, 'postbake'): {
'expectedWidth': 120,
'expectedHeight': 110, # detection params
'transformationTarget':
'cool', # select correct image transformations
'upperKillzone': 550,
'lowerKillzone': 220, # select correct tracking parameters
'rightKillzone': 3000,
'leftKillzone': -3000, # select correct tracking parameters
'timeToDie': 1,
'timeToLive': 0,
'partioningRequired': True
}
}
D = Detector(**param.get((1, 'postbake')))
for i in range(49, 100):
img = cv2.imread("PostBakeSamples/postbake" + str(i) + ".png")
x = D.getImgWithBoxes(img)
while True:
ImgUtils.show("x", x, 0, 0)
key = cv2.waitKey(30)
if key == ord('q'):
break
import time
#infFile = "/home/leo/qj/object_detection/data/socr/inference_frcnn/frozen_inference_graph.pb"
#mapFile = "/home/leo/qj/object_detection/data/socr/pascal_label_map.pbtxt"
infFile = "/home/leo/qj/object_detection/data/socr/inference_ssd/frozen_inference_graph.pb"
mapFile = "/home/leo/qj/object_detection/data/socr/pascal_label_map.pbtxt"
#infFile = "/home/leo/qj/object_detection/data/Index/inference_ssd/frozen_inference_graph.pb"
#mapFile = "/home/leo/qj/object_detection/data/Index/pascal_label_map.pbtxt"
imFiles = [
"/home/leo/qj/object_detection/data/origin/socr_20180410/JPEGImages/im0505.jpg",
"/home/leo/qj/object_detection/data/origin/socr_20180410/JPEGImages/im0506.jpg",
"/home/leo/qj/object_detection/data/origin/socr_20180410/JPEGImages/im0507.jpg",
"/home/leo/qj/object_detection/data/origin/socr_20180410/JPEGImages/im0508.jpg"
]
detector = Detector('socr', infFile, mapFile)
for imFile in imFiles:
img_np = np.asarray(Image.open(imFile))
img_np.setflags(write=1)
s = time.time()
(image_np, tag_boxes, tag_scores, tag_classes,
tag_num) = detector.detect(img_np, visualize=False)
e = time.time()
print 'detecting:', imFile
print 'detect time is:', (e - s)
print 'tag_num:', tag_num
for c in tag_classes:
if tag_scores[c] >= 0.8:
print 'tag:(%s,%f)', c, tag_scores[c]
import argparse
import platform
from detect import Detector
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='Start for face recognition client')
parser.add_argument('--ip', help='Ip of recognition server')
parser.add_argument('--port', help='Port of recognition server')
args = parser.parse_args()
detector = Detector(args.ip, args.port)
if platform.system() == 'Windows' or platform.system() == 'Darwin':
import win_recorder
win_recorder.record_forever(detector)
else:
import pi_recorder
pi_recorder.record_forever(detector)
def detect():
data = request.get_data(cache=False, as_text=False, parse_form_data=False)
provider = BasicAudioProvider(data)
detector = Detector(provider)
return detector.classify()
"""
detector/app.py
Flask-based API webserver to handle inference requests for object detection.
Callum Morrison, 2021
"""
import cv2
import numpy as np
from flask import Flask, Response, abort, jsonify, request
from detect import Detector
app = Flask(__name__)
nn = Detector()
@app.route('/')
def index():
return 'Please use the API at the /api endpoint'
@app.route('/api/detect', methods=['POST'])
def detect():
r = request
array = np.frombuffer(r.data, np.uint8)
img = cv2.imdecode(array, cv2.IMREAD_COLOR)
data = nn.bounding_box(img)
def worker():
provider = ServerAudioProvider(queue)
detector = Detector(provider)
detector.detect()
from detect import Detector
import cv2
detector = Detector(resolution=320)
loop = True
detect_path = 'imgs/img1.jpg'
path, objects, image = detector.detect_objects(detect_path)
cv2.imwrite('detected_image.jpg', image)
image_np = cv2.imread('detected_image.jpg')
cv2.imshow('images', image_np)
cv2.waitKey()
min = distance
f1 = f
print("Face recognised", str(f1[:-4]))
cv2.rectangle(img1, (0, img1.shape[0]), (img1.shape[1], 0),
(0, 255, 0), 3)
font = cv2.FONT_HERSHEY_SIMPLEX
cv2.putText(img1, str(f1[:-4]), (30, img1.shape[0] - 20), font, 1,
(255, 255, 255), 2, cv2.LINE_AA)
cv2.imwrite("result.jpg", img1)
# return [1,str(f[:-4])]
# return [0,""]
obj2 = Detector()
# obj2.detection()
obj1 = Recognition()
obj1.initialise("")
while True:
t1 = threading.Thread(target=obj2.detection())
t2 = threading.Thread(target=obj1.printResults(""))
t1.start()
t2.start()
t1.join()
t2.join()
# while True:
# obj1=Recognition()
# t1.start()
# t1.join()
import cv2
from detect import Detector
from utils import load_config
def initialize_cameras():
cam_1 = cv2.VideoCapture(
f"rtsp://{CAM_USERNAME_1}:{CAM_PASSWORD_1}@{CAM_IP_1}:554/cam/realmonitor?channel=1@subtype=1"
)
return cam_1
config = load_config('configuration.yml')
CAM_IP_1 = config['CAMERA']['IP']['_1']
CAM_USERNAME_1 = config['CAMERA']['USERNAME']['_1']
CAM_PASSWORD_1 = config['CAMERA']['PASSWORD']['_1']
cam_1 = initialize_cameras()
net = Detector()
def main():
yolo = Detector()
# Definition of the parameters
max_cosine_distance = 0.3
nn_budget = None
nms_max_overlap = 1.0
# deep_sort
model_filename = 'model_data/mars-small128.pb'
encoder = gdet.create_box_encoder(model_filename, batch_size=1)
metric = nn_matching.NearestNeighborDistanceMetric("cosine",
max_cosine_distance,
nn_budget)
tracker = Tracker(metric)
writeVideo_flag = True
video_capture = cv2.VideoCapture("town.avi")
if writeVideo_flag:
# Define the codec and create VideoWriter object
w = int(video_capture.get(3))
h = int(video_capture.get(4))
fourcc = cv2.VideoWriter_fourcc(*'MJPG')
out = cv2.VideoWriter('output.avi', fourcc, 15, (w, h))
list_file = open('detection.txt', 'w')
frame_index = -1
fps = 0.0
maxframe = 1
nframe = 0
while True:
ret, frame = video_capture.read() # frame shape 640*480*3
if ret != True:
break
t1 = time.time()
detections = []
# image = Image.fromarray(frame)
nframe += 1
if (nframe >= maxframe):
boxs, obj = yolo.detect(frame)
print(len(boxs))
# print("box_num",len(boxs))
features = encoder(frame, boxs)
# score to 1.0 here).
detections = [
Detection(bbox, 1.0, feature)
for bbox, feature in zip(boxs, features)
]
# Run non-maxima suppression.
boxes = np.array([d.tlwh for d in detections])
scores = np.array([d.confidence for d in detections])
indices = preprocessing.non_max_suppression(
boxes, nms_max_overlap, scores)
detections = [detections[i] for i in indices]
tracker.predict()
tracker.update(detections)
# Call the tracker
nframe = 0
for track in tracker.tracks:
if not track.is_confirmed() or track.time_since_update > 1:
continue
bbox = track.to_tlbr()
# # s=recognize(img)
# try:
# if(len(s)>len(track.track_lpn)):
# track.track_lpn=s
# except Exception as e:
# print(e)
cv2.rectangle(frame, (int(bbox[0]), int(bbox[1])),
(int(bbox[2]), int(bbox[3])), (255, 255, 255), 2)
cv2.putText(frame, str(track.track_id),
(int(bbox[0]), int(bbox[1])), 0, 5e-3 * 200,
(0, 255, 0), 2)
# for det in detections:
# bbox = det.to_tlbr()
# cv2.rectangle(frame,(int(bbox[0]), int(bbox[1])), (int(bbox[2]), int(bbox[3])),(255,0,0), 2)
cv2.imshow('', frame)
if writeVideo_flag:
# save a frame
out.write(frame)
frame_index = frame_index + 1
list_file.write(str(frame_index) + ' ')
fps = (fps + (1. / (time.time() - t1))) / 2
print("fps= %f" % (fps))
# Press Q to stop!
if cv2.waitKey(1) & 0xFF == ord('q'):
break
video_capture.release()
if writeVideo_flag:
out.release()
list_file.close()
cv2.destroyAllWindows()