def batch_detection_example():
args = parser()
check_arguments_errors(args)
batch_size = 3
random.seed(3) # deterministic bbox colors
network, class_names, class_colors = darknet.load_network(
args.config_file,
args.data_file,
args.weights,
batch_size=batch_size
)
image_names = ['data/horses.jpg', 'data/horses.jpg', 'data/eagle.jpg']
images = [cv2.imread(image) for image in image_names]
images, detections, = batch_detection(network, images, class_names,
class_colors, batch_size=batch_size)
for name, image in zip(image_names, images):
cv2.imwrite(name.replace("data/", ""), image)
print(detections)
def __init__(self):
self.image_sub = rospy.Subscriber("/color1/image_raw", Image,
self.callback)
self.pub = rospy.Publisher('/tmp', Int32, queue_size=10)
self.bridge = CvBridge()
r = rospkg.RosPack()
path = r.get_path('yolo') + '/config/libdarknet.so'
self.config_path = r.get_path('yolo') + '/config/yolo-obj.cfg'
self.weight_path = r.get_path('yolo') + '/config/yolo.weights'
self.data_path = r.get_path('yolo') + '/config/obj.data'
self.network, self.class_names, self.class_colors = darknet.load_network(
self.config_path, self.data_path, self.weight_path, batch_size=1)
self.thresh = 0.50
self.width = darknet.network_width(self.network)
self.height = darknet.network_height(self.network)
print('complete to load')
def main():
pub = rospy.Publisher("/yolo_traffic") #WS_ADDED
args = parser()
check_arguments_errors(args)
random.seed(3) # deterministic bbox colors
network, class_names, class_colors = darknet.load_network(
args.config_file,
args.data_file,
args.weights,
batch_size=args.batch_size)
images = load_images(args.input)
index = 0
while True:
# loop asking for new image paths if no list is given
if args.input:
if index >= len(images):
break
image_name = images[index]
else:
image_name = input("Enter Image Path: ")
prev_time = time.time()
image, detections = image_detection(image_name, network, class_names,
class_colors, args.thresh)
# Call Yolo publishing ftn
rt = publish_yolo(detections) #WS_ADDED
if rt != 0:
pub.publish(rt) #WS_ADDED
else:
ROS_INFO("TRAFFIC LIGHT UNDETECTED")
if args.save_labels:
save_annotations(image_name, image, detections, class_names)
darknet.print_detections(detections, args.ext_output)
fps = int(1 / (time.time() - prev_time))
print("FPS: {}".format(fps))
if not args.dont_show:
cv2.imshow('Inference', image)
if cv2.waitKey() & 0xFF == ord('q'):
break
index += 1
def main():
args = parser()
check_arguments_errors(args)
random.seed(3) # deterministic bbox colors
network, class_names, class_colors = darknet.load_network(
args.config_file,
args.data_file,
args.weights,
batch_size=args.batch_size
)
images = load_images(args.input)
index = 0
while True:
# loop asking for new image paths if no list is given
if args.input:
if index >= len(images):
break
image_name = images[index]
else:
image_name = input("Enter Image Path: ")
prev_time = time.time()
image, detections = image_detection(
image_name, network, class_names, class_colors, args.thresh
)
if args.save_labels:
save_annotations(image_name, image, detections, class_names)
darknet.print_detections(detections, args.ext_output)
fps = int(1/(time.time() - prev_time))
print("FPS: {}".format(fps))
#################
#print(darknet.draw_boxes(detections, image_resized, class_colors))
print('Found {} boxes for {}'.format(len(out_boxes), image_name))
if not args.dont_show:
cv2.imshow('Inference', image)
if cv2.waitKey() & 0xFF == ord('q'):
break
index += 1
def main():
args = parser()
check_arguments_errors(args)
random.seed(3) # deterministic bbox colors
network, class_names, class_colors = darknet.load_network(
args.config_file,
args.data_file,
args.weights,
batch_size=args.batch_size
)
print('Load Net Successfully!')
images = load_images(args.input)
vs= VideoStream(usePiCamera=True).start() #
time.sleep(2.0) #
index = 0
while True:
# loop asking for new image paths if no list is given
frame = vs.read() #
frame = imutils.resize(frame, width=400) #
prev_time = time.time()
image, detections = image_detection(
frame, network, class_names, class_colors, args.thresh
)
# if args.save_labels:
save_annotations("image_name.txt", image, detections, class_names)
# darknet.print_detections(detections, args.ext_output)
for label, confidence, bbox in detections:
print(label + ": " + confidence + "%")
fps = int(1/(time.time() - prev_time))
print("FPS: {}".format(fps))
if not args.dont_show:
image = cv2.resize(image, (400, 400))
cv2.imshow('Inference', image)
key = cv2.waitKey(10) & 0xFF
if key == ord('q'):
vs.stop()
break
index += 1
def main():
args = parser()
check_arguments_errors(args)
random.seed(3) # deterministic bbox colors
network, class_names, class_colors = darknet.load_network(
args.config_file,
args.data_file,
args.weights,
batch_size=args.batch_size)
images = load_images(args.input)
index = 0
while True:
# loop asking for new image paths if no list is given
if args.input:
if index >= len(images):
break
image_name = images[index]
name = os.path.basename(image_name)
else:
image_name = input("Enter Image Path: ")
name = image_name.split('/')[-1]
prev_time = time.time()
image, detections = image_detection(image_name, network, class_names,
class_colors, args.thresh)
if args.save_labels:
save_annotations(image_name, image, detections, class_names)
darknet.print_detections(detections, args.ext_output)
path = './prediction/'
if not os.path.exists(path):
os.makedirs(path)
cv2.imwrite(f'{path}/predicted_{name}', image)
fps = int(1 / (time.time() - prev_time))
print("FPS: {}".format(fps))
#if not args.dont_show:
# cv2.imshow('Inference', image)
# if cv2.waitKey() & 0xFF == ord('q'):
# break
index += 1
def InitialiseYOLO():
print('YOLO Init')
global network, class_names, darknet_image
#configPath = "./cfg/yolov4.cfg" /home/yokoyamalab/tmp/darknet/cfg/person.cfg
configPath = "/home/hayashida/yolov4/darknet/cfg/yolov4.cfg"
#weightPath = "./yolov4.weights" /home/yokoyamalab/tmp/darknet/backup/person/mix2/person_best.weights
weightPath = "/home/hayashida/yolov4/darknet/yolov4.weights"
#metaPath = "./cfg/coco.data" /home/yokoyamalab/tmp/darknet/cfg/person.data
dataPath = "/home/hayashida/yolov4/darknet/cfg/coco.data"
network, class_names, class_colors = darknet.load_network(configPath,
dataPath,
weightPath,
batch_size=1)
# Darknet doesn't accept numpy images.
# Create one with image we reuse for each detect
width = darknet.network_width(network)
height = darknet.network_height(network)
darknet_image = darknet.make_image(width, height, 3)
return network, class_names, class_colors
def main(
cfg = "./cfg/yolov4-tiny.cfg",
data = "./cfg/coco.data",
weights = "./bin/yolov4-tiny.weights"
):
global network, class_names, class_colors, width, height, darknet_image
network, class_names, class_colors = darknet.load_network(
cfg,
data,
weights,
batch_size=1
)
width = darknet.network_width(network)
height = darknet.network_height(network)
darknet_image = darknet.make_image(width, height, 3)
width = int(width)
height = int(height)
print(width, height)
ip = "192.168.55.1"
app.run(host=ip, port=5000, debug=False)
def __init__(self, detect_config):
print("=== {}".format(YOLOv4ObjectDetector))
parser = DetectConfigParser(detect_config)
self.WEIGHTS = parser.weights()
self.CONFIG = parser.config()
self.DATA = parser.data()
self.OUTPUTS = parser.outputs()
self.THRESHOLD = parser.threshold()
self.SAVE_LABELS = True
self.output_dir = os.path.join(os.getcwd(), self.OUTPUTS)
if not os.path.exists(self.output_dir):
os.makedirs(self.output_dir)
#random.seed(3) # deterministic bbox color
self.BATCH_SIZE = 1
self.network, self.class_names, self.class_colors = darknet.load_network(
self.CONFIG, self.DATA, self.WEIGHTS, batch_size=self.BATCH_SIZE)
print("=== class_names {}".format(self.class_names))
def main():
args = parser()
check_arguments_errors(args)
random.seed(3) # deterministic bbox colors
network, class_names, class_colors = darknet.load_network(
args.config_file,
args.data_file,
args.weights,
batch_size=args.batch_size)
images = load_images(args.input)
index = 0
while True:
# loop asking for new image paths if no list is given
if args.input:
if index >= len(images):
break
image_name = images[index]
else:
image_name = input("Enter Image Path: ")
prev_time = time.time()
image, detections = image_detection(image_name, network, class_names,
class_colors, args.thresh)
if args.save_labels:
save_annotations(image_name, image, detections, class_names)
darknet.print_detections(detections, args.ext_output)
fps = int(1 / (time.time() - prev_time))
print("FPS: {}".format(fps))
cv2.imwrite(
os.path.join(args.output, "out_" + image_name.split("/")[-1]),
image)
if not args.dont_show:
im = Image.open(
os.path.join(args.output, "out_" + image_name.split("/")[-1]))
im.show()
if cv2.waitKey() & 0xFF == ord('q'):
break
index += 1
def __init__(self):
self.network, self.class_names, self.class_colors = darknet.load_network(
config_file="./cfg/yolov3-voc-mask-detect.cfg",
data_file="./data/voc-mask-detect.data",
weights="./backup/yolov3-voc-mask-detect_final-extend.weights",
batch_size=1)
self.mask_detection = MaskDetection()
self.drew_images = Queue()
self.input_video = "sources/mask-detect-test.mp4"
self.output_video = None
self.frame_queue = Queue()
self.frame_queue_face = Queue()
self.face_image_queue = Queue(maxsize=1)
self.darknet_image_queue = Queue(maxsize=1)
self.detections_queue = Queue(maxsize=1)
self.face_detections_queue = Queue(maxsize=1)
self.fps_queue = Queue(maxsize=1)
self.width = darknet.network_width(self.network)
self.height = darknet.network_height(self.network)
self.cap = cv2.VideoCapture(0)
self.darknet_image = darknet.make_image(self.width, self.height, 3)
self.thresh = 0.5
def main():
args = parser()
check_arguments_errors(args)
random.seed(3) # deterministic bbox colors
network, class_names, class_colors = darknet.load_network(
args.config_file,
args.data_file,
args.weights,
batch_size=args.batch_size
)
images = load_images(args.input)
index = 0< /home/fadi/다운로드/Yolo_mark/x64/Release/data/train.txt >
while True:
# loop asking for new image paths if no list i< /home/fadi/다운로드/Yolo_mark/x64/Release/data/train.txt >s given
if args.input:
if index >= len(images):
break
image_name = images[index]
else:
image_name = input("Enter Image Path: ")
prev_time = time.time()
image, detections = image_detection(
image_name, network, class_names, class_colors, args.thresh
)< /home/fadi/다운로드/Yolo_ /home/fadi/다운로드/Yolo_mark/x64/Release/data/train.txt mark/x64/Release/data/train.txt >
if args.save_labels:
save_annotations(image_name, image, detections, class_names)
darknet.print_detections(detections, args.ext_output)
fps = int(1/(time.time() - prev_time))
print("FPS: {}".format(fps))
if not args.dont_show:
cv2.imshow('Inference', image)
#cv2.imwrite('./'+str(index)+'.jpg', image)
if cv2.waitKey() & 0xFF == ord('q'):
break
index += 1
def batch_detection_example():
args = parser()
check_arguments_errors(args)
batch_size = 64
random.seed(64) # deterministic bbox colors
network, class_names, class_colors = darknet.load_network(
args.config_file, args.data_file, args.weights, batch_size=batch_size)
image_names = [
'/ws/data/ADD/images/2_23_000020_097.jpg',
'/ws/data/ADD/images/2_23_000020_097.jpg',
'/ws/data/ADD/images/2_23_000020_097.jpg',
'/ws/data/ADD/images/3_35_000008_318.jpg',
'/ws/data/ADD/images/4_49_000031_838.jpg'
]
images = [cv2.imread(image) for image in image_names]
images, detections, = batch_detection(network,
images,
class_names,
class_colors,
batch_size=batch_size)
for name, image in zip(image_names, images):
cv2.imwrite(name.replace("data/", ""), image)
print(detections)
def build_detection_engine(config_file: str, data_file: str, weights: str,
thresh: float):
network, class_names, class_colors = darknet.load_network(config_file,
data_file,
weights,
batch_size=1)
labels_rev = {}
for i, label in enumerate(class_names):
labels_rev[label] = i
width = darknet.network_width(network)
height = darknet.network_height(network)
darknet_image = darknet.make_image(width, height, 3)
def inference(flow_id: str, frame: object):
image = frame_data_2_bytes(frame, width, height)
darknet.copy_image_from_bytes(darknet_image, image)
detections = darknet.detect_image(network,
class_names,
darknet_image,
thresh=thresh)
result = PyDetectionBox(frame_id=frame.frame_id, engine_id='darknet')
for label, confidence, bbox in detections:
left, top, right, bottom = darknet.bbox2points(bbox)
result.add_box(category_id=labels_rev.get(label, ''),
category_label=label,
x1=left,
y1=top,
x2=right,
y2=left,
probability=float(confidence))
darknet.free_image(darknet_image)
return flow_id, result
return inference
name = ''
if idx_track in idx_dict:
name = idx_dict[idx_track]['name']
cv2.rectangle(image, (left, top), (right, bottom), (255, 0, 0), 1)
cv2.putText(image, "ID: {} {}".format(idx_track,
name), (left, top - 5),
cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)
return image
if __name__ == '__main__':
db_dir = './database'
MAX_FRAME = 15
THRESHOLD = 1.0
# load yolo
network, class_names, colors = darknet.load_network(
"yolo/fs_192.cfg", "yolo/obj.data", "yolo/yolo.weights")
vid_path = 0
#vid_path = "videoplayback.mp4"
cap = cv2.VideoCapture(vid_path)
vid_width = int(cap.get(3))
vid_height = int(cap.get(4))
vid_fps = cap.get(5)
width = darknet.network_width(network)
height = darknet.network_height(network)
darknet_image = darknet.make_image(width, height, 3)
mot_tracker = Sort()
# load face feature extraction model
export_dir = "./tf131_model"
sess = tf.Session(graph=tf.Graph())
tf.saved_model.loader.load(sess, [tf.saved_model.tag_constants.SERVING],
export_dir)
help='Show verbose finded and processed data')
parser.add_argument('-s',
'--save',
action='store_true',
required=False,
help='Save objects to directory')
args = parser.parse_args()
if (args.image is None and args.video is None):
print('No input given!')
sys.exit(-1)
configPath = './cfg/yolov4.cfg'
weightPath = './cfg/yolov4.weights'
metaPath = './cfg/coco.data'
net, classes, colors = darknet.load_network(configPath, metaPath, weightPath)
# Single image processing
if (args.image is not None):
filename = args.image
# Read image image
im = cv2.imread(filename)
imheight, imwidth, channels = im.shape
# Create an image we reuse for each detect
darknet_image = darknet.make_image(imwidth, imheight, channels)
darknet.copy_image_from_bytes(darknet_image, im.tobytes())
detections = darknet.detect_image(net, classes, darknet_image)
print(detections)
import argparse
import cv2
import youtube_dl
import darknet
import ultis
import caches
parser = argparse.ArgumentParser()
parser.add_argument("--url", help="Live youtube url")
parser.add_argument("--quality", help="quality of video", default=720)
args = parser.parse_args()
darknet_network, class_names, _ = darknet.load_network(
"./cfg/vizyal.cfg",
"./cfg/vizyal.data",
"./models/vizyal.weights",
batch_size=1
)
caches.darknet_width = darknet.network_width(darknet_network)
caches.darknet_height = darknet.network_height(darknet_network)
if __name__ == '__main__':
ydl_opts = {}
ydl = youtube_dl.YoutubeDL(ydl_opts)
info_dict = ydl.extract_info(args.url, download=False)
formats = info_dict.get('formats', None)
frame_id = 0
for f in formats:
if (f.get('height', None) == int(args.quality)):
import cv2
from flask import Flask, request, Response
import darknet
import os
from PIL import Image
from io import BytesIO
from flask_cors import CORS
app = Flask(__name__)
CORS(app)
app.config["DEBUG"] = True
app.config["IMAGE_UPLOADS"] = "./upload/"
network, class_names, class_colors = darknet.load_network(
"./custom-yolov4-detector.cfg",
"./obj.data",
"./custom-yolov4-detector_final_1.weights",
batch_size=1)
def image_detection(image_path, network, class_names, class_colors, thresh):
# Darknet doesn't accept numpy images.
# Create one with image we reuse for each detect
width = darknet.network_width(network)
height = darknet.network_height(network)
darknet_image = darknet.make_image(width, height, 3)
image = cv2.imread(image_path)
image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
image_resized = cv2.resize(image_rgb, (width, height),
interpolation=cv2.INTER_LINEAR)
if 'out.mp4' != None:
video.write(image)
if cv2.waitKey(fps) == 27:
break
cap.release()
video.release()
cv2.destroyAllWindows()
if __name__ == '__main__':
frame_queue = Queue()
darknet_image_queue = Queue(maxsize=1)
detections_queue = Queue(maxsize=1)
fps_queue = Queue(maxsize=1)
# args = parser()
# check_arguments_errors(args)
network, class_names, class_colors = darknet.load_network(
'yolov4-tiny-obj.cfg',
'obj.data',
'yolov4-tiny-obj_best.weights',
batch_size=1
)
darknet_width = darknet.network_width(network)
darknet_height = darknet.network_height(network)
# input_path = str2int(args.input)
input_path = "1.mp4"
cap = cv2.VideoCapture(input_path)
Thread(target=video_capture, args=(frame_queue, darknet_image_queue)).start()
Thread(target=inference, args=(darknet_image_queue, detections_queue, fps_queue)).start()
Thread(target=drawing, args=(frame_queue, detections_queue, fps_queue)).start()
if args.out_filename is not None:
video.release()
cv2.destroyAllWindows()
return
if __name__ == '__main__':
frame_queue = Queue()
darknet_image_queue = Queue(maxsize=1)
detections_queue = Queue(maxsize=1)
fps_queue = Queue(maxsize=1)
args = parser()
check_arguments_errors(args)
network, class_names, class_colors = darknet.load_network(args.config_file,
args.data_file,
args.weights,
batch_size=1)
# Darknet doesn't accept numpy images.
# Create one with image we reuse for each detect
width = darknet.network_width(network)
height = darknet.network_height(network)
darknet_image = darknet.make_image(width, height, 3)
input_path = str2int(args.input)
cap = cv2.VideoCapture(input_path)
Thread(target=video_capture,
args=(frame_queue, darknet_image_queue)).start()
Thread(target=inference,
args=(darknet_image_queue, detections_queue, fps_queue)).start()
Thread(target=drawing,
args=(frame_queue, detections_queue, fps_queue)).start()
def YOLO():
"""
Perform Object detection
"""
global metaMain, netMain, altNames
configPath = "./cfg/yolov4.cfg"
weightPath = "./yolov4.weights"
metaPath = "./cfg/coco.data"
deep_sort = "./deep_sort_app.py"
if not os.path.exists(configPath):
raise ValueError("Invalid config path `" +
os.path.abspath(configPath) + "`")
if not os.path.exists(weightPath):
raise ValueError("Invalid weight path `" +
os.path.abspath(weightPath) + "`")
if not os.path.exists(metaPath):
raise ValueError("Invalid data file path `" +
os.path.abspath(metaPath) + "`")
if netMain is None:
network, class_names, class_colors = darknet.load_network(configPath,
metaPath,
weightPath,
batch_size=1)
# netMain = darknet.load_net_custom(configPath.encode(
# "ascii"), weightPath.encode("ascii"), 0, 1) # batch size = 1
if metaMain is None:
metaMain = darknet.load_meta(metaPath.encode("ascii"))
if altNames is None:
try:
with open(metaPath) as metaFH:
metaContents = metaFH.read()
import re
match = re.search("names *= *(.*)$", metaContents,
re.IGNORECASE | re.MULTILINE)
if match:
result = match.group(1)
else:
result = None
try:
if os.path.exists(result):
with open(result) as namesFH:
namesList = namesFH.read().strip().split("\n")
altNames = [x.strip() for x in namesList]
except TypeError:
pass
except Exception:
pass
#cap = cv2.VideoCapture(0)
cap = cv2.VideoCapture("./data/bridge.mp4")
frame_width = int(cap.get(3))
frame_height = int(cap.get(4))
new_height, new_width = frame_height // 2, frame_width // 2
# print("Video Reolution: ",(width, height))
out = cv2.VideoWriter("./demo/traffic_output.avi",
cv2.VideoWriter_fourcc(*"MJPG"), 10.0,
(new_width, new_height))
# print("Starting the YOLO loop...")
# Create an image we reuse for each detect
darknet_image = darknet.make_image(new_width, new_height, 3)
while True:
prev_time = time.time()
ret, frame_read = cap.read()
# Check if frame present :: 'ret' returns True if frame present, otherwise break the loop.
if not ret:
break
frame_rgb = cv2.cvtColor(frame_read, cv2.COLOR_BGR2RGB)
frame_resized = cv2.resize(frame_rgb, (new_width, new_height),
interpolation=cv2.INTER_LINEAR)
darknet.copy_image_from_bytes(darknet_image, frame_resized.tobytes())
# detections = darknet.detect_image(netMain, metaMain, darknet_image, thresh=0.25)
detections = darknet.detect_image(network,
class_names,
darknet_image,
thresh=0.25)
image = cvDrawBoxes(detections, frame_resized)
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
print(1 / (time.time() - prev_time))
cv2.imshow('Demo', image)
cv2.waitKey(3)
out.write(image)
cap.release()
out.release()
print(":::Video Write Completed")
restore_var = [v for v in tf.global_variables()]
loader = tf.train.Saver(var_list=restore_var)
ckpt = const.SNAPSHOT_DIR
load(loader, sess, ckpt)
# Yolo Initializer
config_file = "/content/test/cfg/yolov4.cfg"
weights = "/content/test/yolov4.weights"
data_file = "/content/test/cfg/coco.data"
batch_size = 1
thresh = 0.25 # remove object with confidence score < 0.25
save_labels = False # do not export output to .txt file
random.seed(3) # deterministic bbox colors
network, class_names, class_colors = darknet.load_network(
config_file, data_file, weights, batch_size=batch_size)
count = 0
# print("Images: ", images)
# Initialize MONAD
# Create Training Feature Space
Train_FS = list()
trainingClasses = ["person"]
import json
with open('result_train_ucsd.json', 'r') as f:
D = json.load(f)
train_objects = []
for i in range(len(D)):
def YOLO():
global altNames
configPath = "../cfg/yolov4.cfg" # Path to cfg
weightPath = "../yolov4.weights" # Path to weights
metaPath = "../cfg/coco.data" # Path to meta data
if not os.path.exists(
configPath
): # Checks whether file exists otherwise return ValueError
raise ValueError("Invalid config path `" +
os.path.abspath(configPath) + "`")
if not os.path.exists(weightPath):
raise ValueError("Invalid weight path `" +
os.path.abspath(weightPath) + "`")
if not os.path.exists(metaPath):
raise ValueError("Invalid data file path `" +
os.path.abspath(metaPath) + "`")
network, class_names, class_colors = darknet.load_network(configPath,
metaPath,
weightPath,
batch_size=1)
if altNames is None:
try:
# Read the metadata file and find the class names file path.
with open(metaPath) as metaFH:
metaContents = metaFH.read()
import re
match = re.search("names *= *(.*)$", metaContents,
re.IGNORECASE | re.MULTILINE)
if match:
result = match.group(1)
else:
result = None
try:
if os.path.exists(result):
with open(result) as namesFH:
namesList = namesFH.read().strip().split("\n")
altNames = [x.strip() for x in namesList]
except TypeError:
pass
except Exception:
pass
#cap = cv2.VideoCapture(0) # Uncomment to use Webcam
cap = cv2.VideoCapture(
"../data/test50.mp4") # Local Stored video detection - Set input video
#cap = cv2.VideoCapture("../data/Video_for_Testing.mp4") # Local Stored video detection - Set input video
frame_width = int(
cap.get(3)) # Returns the width and height of capture video
frame_height = int(cap.get(4))
# Set out for video writer
out = cv2.VideoWriter( # Set the Output path for video writer
"./Demo/output.avi", cv2.VideoWriter_fourcc(*"MJPG"), 10.0,
(frame_width, frame_height))
print("Starting the YOLO loop...")
# Create an image we reuse for each detect
darknet_image = darknet.make_image(
frame_width, frame_height,
3) # Create image according darknet for compatibility of network
while True: # Load the input frame and write output frame.
prev_time = time.time()
ret, frame_read = cap.read(
) # Capture frame and return true if frame present
# For Assertion Failed Error in OpenCV
if not ret: # Check if frame present otherwise he break the while loop
break
frame_rgb = cv2.cvtColor(
frame_read, cv2.COLOR_BGR2RGB
) # Convert frame into RGB from BGR and resize accordingly
frame_resized = cv2.resize(frame_rgb, (frame_width, frame_height),
interpolation=cv2.INTER_LINEAR)
darknet.copy_image_from_bytes(
darknet_image,
frame_resized.tobytes()) # Copy that frame bytes to darknet_image
detections = darknet.detect_image(
network, class_names, darknet_image,
thresh=0.25) # Detection occurs at this line and return
# detections, for customize we can change the threshold.
image = cvDrawBoxes(
detections, frame_resized
) # Call the function cvDrawBoxes() for colored bounding box per class
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
print(1 / (time.time() - prev_time))
cv2.imshow('Demo', image) # Display Image window
cv2.waitKey(3)
out.write(image) # Write that frame into output video
cap.release() # For releasing cap and out.
out.release()
print(":::Video Write Completed")
def __init__(self, cam_num, parent=None):
self.cam_num = cam_num
#載入資料庫的資料
self.reload()
QtWidgets.QMainWindow.__init__(self)
#self.setWindowTitle("ggez")
self.ui = uic.loadUi("webcam.ui", self)
#self.ui.setFixedSize(self.size())
self.ui.tabWidget.setTabText(0, "Main")
self.ui.tabWidget.setTabText(1, "Search")
self.ui.tabWidget.setTabText(2, "Setting")
self.ui.Open_Button.clicked.connect(self.open_detect)
self.ui.pushButton.clicked.connect(self.save_image)
self.ui.Stop_Button.clicked.connect(self.stop)
self.ui.lineEdit.returnPressed.connect(self.save_image)
self.ui.Send_Button.clicked.connect(self.search_event)
self.ui.lineEdit_2.returnPressed.connect(self.search_event)
#由資料庫抓取資料
self.ui.reload_image.clicked.connect(self.reload)
self.ui.Recognition_checkbox.clicked.connect(
self.Recognition_check_event)
self.Recognition_checkbox.setStyleSheet(
"QPushButton{background:black;border-radius:20;}")
self.widget.setStyleSheet(
"QWidget{border-width:2px;}"
"QWidget{border-color:black;}"
"QWidget{border-style:outset;}"
"QWidget{height:100;}"
"QWidget{border-radius:5px;}"
"QWidget{background-color:qlineargradient(x1 : 0, y1 : 0, x2 : 0, y2 : 1, stop : 0.0 #f5f9ff,stop : 0.5 #c7dfff,stop : 0.55 #afd2ff,stop : 1.0 #c0dbff);}"
)
self.ui.close_button.clicked.connect(self.close_camera)
self.ui.tabWidget.setTabIcon(0, QtGui.QIcon("home.png"))
self.ui.tabWidget.setIconSize(QtCore.QSize(30, 30))
self.ui.tabWidget.setTabIcon(1, QtGui.QIcon("search.png"))
self.ui.tabWidget.setIconSize(QtCore.QSize(30, 30))
self.ui.tabWidget.setTabIcon(2, QtGui.QIcon("setting.png"))
self.ui.tabWidget.setIconSize(QtCore.QSize(30, 30))
self.setWindowIcon(QtGui.QIcon("window_icon.png"))
tab_shape = QtWidgets.QTabWidget.Triangular
self.tabWidget.setTabShape(tab_shape)
self.check = 0
self.i = 0
self.in_or_out = 0
self.Recognition_check = False
self.label2 = QLabel(self)
self.label1 = QLabel(self)
self.label1.setStyleSheet(
"QLabel{background:0;}"
"QLabel{color:rgb(300,300,300,120);font-size:30px;font-weight:bold;font-family:宋体;}"
)
# 動態顯示時間在label上
timer = QTimer(self)
timer.timeout.connect(self.showtime)
timer.start()
self.label1.setAlignment(QtCore.Qt.AlignCenter)
self.label2.setGeometry(self.label.width() + 20, 150, 45, 45)
#self.label1.resize(self.width(),50)
# In[]
#----------------口罩辨識初始化-----------------
self.frame_queue = Queue()
self.darknet_image_queue = Queue(maxsize=1)
self.detections_queue = Queue(maxsize=1)
self.fps_queue = Queue(maxsize=1)
#宣告一個放照片的Queue
self.YOLO_image_queue = Queue(maxsize=1)
#宣告一個放人臉的Queue
self.YOLO_face_queue = Queue(maxsize=1)
self.face = 0
self.recorded_people = []
self.clock = 0
self.args = parser()
check_arguments_errors(self.args)
self.network, self.class_names, self.class_colors = darknet.load_network(
self.args.config_file,
self.args.data_file,
self.args.weights,
batch_size=1)
# Darknet doesn't accept numpy images.
# Create one with image we reuse for each detect
self.d_width = darknet.network_width(self.network)
self.d_height = darknet.network_height(self.network)
self.darknet_image = darknet.make_image(self.d_width, self.d_height, 3)
self.show()
h_scale = float(t_h) / s_h
res = []
for label, confidence, bbox in detections:
x, y, w, h = bbox
x = x * w_scale
y = y * h_scale
w = w * w_scale
h = h * h_scale
left, top, right, bottom = darknet.bbox2points((x, y, w, h))
res.append((left, top, right - left, bottom - top, confidence))
return res
if __name__ == '__main__':
network, class_names, colors = darknet.load_network(
"yolo/yolov4-tiny-3l.cfg", "yolo/obj.data",
"yolo/yolov4-tiny-3l_best.weights")
width = darknet.network_width(network)
height = darknet.network_height(network)
darknet_image = darknet.make_image(width, height, 3)
#
# frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
# frame_resized = cv2.resize(frame_rgb, (width, height),
# interpolation=cv2.INTER_LINEAR)
# darknet.copy_image_from_bytes(darknet_image, frame_resized.tobytes())
# detections = darknet.detect_image(network, class_names, darknet_image, thresh=0.3)
# frame = draw_boxes(detections, frame, (width, height))
val_dir = '/home/khang/Khang/data/wider/val/WIDER_val/images'
dirs = []
for dir_name in os.listdir(val_dir):
if os.path.isdir(os.path.join(val_dir, dir_name)):
def YOLO():
# global metaMain, netMain, altNames
configPath = "./cfg/knockknock_cfg.cfg" # Path to cfg
weightPath = "./knockknock_cfg_best.weights" # Path to weights
metaPath = "./data/obj.data" # Path to meta data
if not os.path.exists(
configPath
): # Checks whether file exists otherwise return ValueError
raise ValueError("Invalid config path `" +
os.path.abspath(configPath) + "`")
if not os.path.exists(weightPath):
raise ValueError("Invalid weight path `" +
os.path.abspath(weightPath) + "`")
if not os.path.exists(metaPath):
raise ValueError("Invalid data file path `" +
os.path.abspath(metaPath) + "`")
network, class_names, class_colors = darknet.load_network(configPath,
metaPath,
weightPath,
batch_size=1)
#cap = cv2.VideoCapture(0) # Uncomment to use Webcam
cap = cv2.VideoCapture(
"ups02.avi") # Local Stored video detection - Set input video
frame_width = int(
cap.get(3)) # Returns the width and height of capture video
frame_height = int(cap.get(4))
# Set out for video writer
out = cv2.VideoWriter( # Set the Output path for video writer
"./output.avi", cv2.VideoWriter_fourcc(*"MJPG"), 10.0,
(frame_width, frame_height))
print("Starting the YOLO loop...")
# Create an image we reuse for each detect
darknet_image = darknet.make_image(
frame_width, frame_height,
3) # Create image according darknet for compatibility of network
x = float(0)
while True: # Load the input frame and write output frame.
prev_time = time.time()
# print("line 107 pass")
ret, frame_read = cap.read(
) # Capture frame and return true if frame present
# print("line 109 pass")
# For Assertion Failed Error in OpenCV
if not ret: # Check if frame present otherwise he break the while loop
break
frame_rgb = cv2.cvtColor(
frame_read, cv2.COLOR_BGR2RGB
) # Convert frame into RGB from BGR and resize accordingly
#print("line 116 pass")
frame_resized = cv2.resize(frame_rgb, (frame_width, frame_height),
interpolation=cv2.INTER_LINEAR)
#print("line 117 pass")
darknet.copy_image_from_bytes(
darknet_image,
frame_resized.tobytes()) # Copy that frame bytes to darknet_image
#print("line 121 pass")
# detections = darknet.detect_image(netMain, metaMain, darknet_image, thresh=0.5) # Detection occurs at this line and return detections, for customize we can change the threshold.
detections = darknet.detect_image(network,
class_names,
darknet_image,
thresh=0.8)
#print("line 124 pass")
image = cvDrawBoxes(
detections, frame_resized
) # Call the function cvDrawBoxes() for colored bounding box per class
#print("line 126 pass")
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
# if detections:
# print(detections[0][1])
if detections:
if float(detections[0][1]) > x:
snap = image
detection = detections
x = float(detections[0][1])
# x=x+1
# print(detections)
#print("line 128 pass")
# print(1/(time.time()-prev_time))
#print("line 130 pass")
#cv2.imshow('Demo', image) # Display Image window
#print("line 132 pass")
cv2.waitKey(3)
out.write(image) # Write that frame into output video
# print("detections[0]: " + str(detections[0][0]))
cv2.imwrite("snap.jpg", snap)
cap.release() # For releasing cap and out.
out.release()
print(detection[0][0])
print(detection[0][1])
print(":::Video Write Completed")
from pandas import DataFrame
import darknet
import sys
import itertools
from tqdm import tqdm
sys.path.insert(
0, '/mekeneocr/myUniverse/darknet'
) #puth darknet path in first place in sys (this step is very important)
config = "/mekeneocr/tiny/yolov4-tiny-custom.cfg"
data = "/mekeneocr/tiny/letters.data"
weights = "/mekeneocr/tiny/yolov4-tiny-custom_best.weights"
network, class_names, class_colors = darknet.load_network(config,
data,
weights,
batch_size=1)
def image_detection(image_path, network, class_names, class_colors, thresh):
# Darknet doesn't accept numpy images.
# Create one with image we reuse for each detect
width = darknet.network_width(network)
height = darknet.network_height(network)
darknet_image = darknet.make_image(width, height, 3)
image = cv2.imread(image_path)
image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
image_resized = cv2.resize(image_rgb, (width, height),
interpolation=cv2.INTER_LINEAR)
label = sorted(label)
label = str(label) + get_direction()
cv.putText(image, str(label), (5, 40), FONT_SIMPLEX, 1, COLOR_B, 2, LINE_AA)
return image
if __name__ == '__main__':
# Initialization
names_path = '../config/forklift.names'
cfg_path = '../config/forklift.cfg'
data_path = '../config/forklift.data'
weighs_path = '../weights/forklift.weights'
video_path = '../videos/forklift.mp4'
threshold = 0.5
# Initialize yolov4 network
net, names, colors = darknet.load_network(cfg_path, data_path, weighs_path, 1)
winName = 'Deep learning object detection in YOLO'
cv.namedWindow(winName, cv.WINDOW_NORMAL)
cv.resizeWindow(winName, IMAGE_WIDTH, IMAGE_HEIGHT)
cv.moveWindow(winName, 100, 100)
while True:
# Initialization
Target_Objections = {'forklift': Attributes(), 'cargo': Attributes()}
cap = cv.VideoCapture(video_path)
while cv.waitKey(1) < 0:
state, frame = cap.read()
if not state:
break
# Detect
detections = image_detection(frame, net, names, threshold)
# Draw trace
def main():
# YOLO and darknetconfiguration
configPath = "./cfg/knockknock_cfg.cfg" # Path to cfg
weightPath = "./knockknock_cfg_best.weights" # Path to weights
metaPath = "./data/obj.data" # Path to meta data
if not os.path.exists(
configPath
): # Checks whether file exists otherwise return ValueError
raise ValueError("Invalid config path `" +
os.path.abspath(configPath) + "`")
if not os.path.exists(weightPath):
raise ValueError("Invalid weight path `" +
os.path.abspath(weightPath) + "`")
if not os.path.exists(metaPath):
raise ValueError("Invalid data file path `" +
os.path.abspath(metaPath) + "`")
network, class_names, class_colors = darknet.load_network(configPath,
metaPath,
weightPath,
batch_size=1)
# rtsp = rtsp://admin:[email protected]:554
#cap = cv2.VideoCapture(rtsp) To run on camera
#cap = cv2.VideoCapture(0) # Uncomment to use Webcam
cap = cv2.VideoCapture(
videoInput) # Local Stored video detection - Set input video
frame_width = int(
cap.get(3)) # Returns the width and height of capture video
frame_height = int(cap.get(4))
# Set out for video writer
fileName = datetime.now().strftime("%B-%d-%y_%H:%M:%S")
outputPath = "./results/" + fileName + ".avi"
out = cv2.VideoWriter( # Set the Output path for video writer
outputPath, cv2.VideoWriter_fourcc(*"MJPG"), 10.0,
(frame_width, frame_height))
print("Analyze starts..")
darknet_image = darknet.make_image(
frame_width, frame_height,
3) # Create image according darknet for compatibility of network
x = float(0)
while True: # Load the input frame and write output frame.
prev_time = time.time()
# print("line 107 pass")
ret, frame_read = cap.read(
) # Capture frame and return true if frame present
# print("line 109 pass")
# For Assertion Failed Error in OpenCV
if not ret: # Check if frame present otherwise he break the while loop
break
frame_rgb = cv2.cvtColor(
frame_read, cv2.COLOR_BGR2RGB
) # Convert frame into RGB from BGR and resize accordingly
#print("line 116 pass")
frame_resized = cv2.resize(frame_rgb, (frame_width, frame_height),
interpolation=cv2.INTER_LINEAR)
#print("line 117 pass")
darknet.copy_image_from_bytes(
darknet_image,
frame_resized.tobytes()) # Copy that frame bytes to darknet_image
#print("line 121 pass")
# detections = darknet.detect_image(netMain, metaMain, darknet_image, thresh=0.5) # Detection occurs at this line and return detections, for customize we can change the threshold.
detections = darknet.detect_image(network,
class_names,
darknet_image,
thresh=0.8)
#print("line 124 pass")
image = drawBoxes(
detections, frame_resized
) # Call the function cvDrawBoxes() for colored bounding box per class
#print("line 126 pass")
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
if detections:
if float(detections[0][1]) > x:
snap = image
detection = detections
x = float(detections[0][1])
# print(1/(time.time()-prev_time))
#cv2.imshow('Demo', image) # Display Image window
cv2.waitKey(3)
out.write(image) # Write that frame into output video
# print("detections[0]: " + str(detections[0][0]))
cap.release() # For releasing cap and out.
out.release()
if x > float(80):
snapPath = "./results/" + fileName + ".jpg"
cv2.imwrite(snapPath, snap)
label = str(detection[0][0])
confidence = float(detection[0][1])
logFilePath = logging(label, confidence)
sendAttachments(snapPath, logFilePath, label, confidence)
print("Analyze Completed")
import darknet
import time
# Parameters
win_title = 'YOLOv4 CUSTOM DETECTOR'
cfg_file = 'cfg/yolov4-tiny.cfg'
data_file = 'cfg/coco.data'
weight_file = 'yolov4-tiny.weights'
thre = 0.25
show_coordinates = False
show_size = (800, 800)
# Load Network
network, class_names, class_colors = darknet.load_network(
cfg_file,
data_file,
weight_file,
batch_size=1
)
# Get Nets Input dimentions
width = darknet.network_width(network)
height = darknet.network_height(network)
# Get Webcam from cv2
cap = cv2.VideoCapture(0)
# Video Stream
while cap.isOpened():
# Get current frame, quit if no frame
ret, frame = cap.read()
