def main():
sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
sock.connect(('localhost', 51000))
detecter = Detection()
dp = Display()
ad = Audio()
thread = Thread(target=dp.start)
thread.setDaemon(True)
thread.start()
state_before = 0
while True:
motor_power, target_dist = detecter.detect()
# 走行プログラムと送信
sock.sendall(motor_power.to_bytes(2, 'big') + target_dist.to_bytes(2, 'big'))
byte_data = sock.recv(4)
#print(byte_data)
state = int.from_bytes(byte_data[:2], 'big')
ad_flag = int.from_bytes(byte_data[2:], 'big')
#print(state)
if state != state_before:
dp.changeImage(state)
if ad_flag == 1:
if state_before == 0:
ad.play(0)
elif state == 5:
ad.play(1)
elif state_before == 5:
ad.play(2)
state_before = state
# print(frame.shape)
count += 1
if count % 100 == 0:
print(basename + " progress: ", count)
if lost_frame_count > 20:
# Global detector to the rescue
img_size = max(frame.shape[0], frame.shape[1])
x_detection = frame.transpose(2, 0, 1)
# print(x_detection.shape)
detection_result = detector.detect(np.squeeze(x_detection))
# cv2.imwrite("test.jpg", np.squeeze(x_crop).transpose((1,2,0)))
dboxes = detection_result["instances"].pred_boxes.tensor.cpu(
).detach().numpy()
dscores = detection_result["instances"].scores.cpu().detach(
).numpy()
dboxes = dboxes.astype(np.int)
if len(dscores) == 0:
out.write(frame)
continue
def window_penalty(dboxes, dscores, center_pos, img_size):
new_score = []
for idx in range(len(dscores)):
cv2.circle(frame, (int(tracker.tracker.center_pos[0]), int(tracker.tracker.center_pos[1])), 7, (0, 0, 255), -1)
x_crop, scale_z = tracker.get_roi(frame, search_instance_size)
# if count == 1338:
# cv2.imwrite("result.jpg", np.squeeze(x_crop).transpose(1, 2, 0))
# print(x_crop.shape, scale_z)
center_pos = tracker.tracker.center_pos
search_region = [(center_pos - search_instance_size/scale_z/2), (center_pos + search_instance_size/scale_z/2)]
# print(search_region)
drawrect(frame, (int(search_region[0][0]), int(search_region[0][1])),
(int(search_region[1][0]), int(search_region[1][1])),
(0, 255, 0), 2)
# print(x_crop.shape, scale_z)
detection_result = detector.detect(np.squeeze(x_crop))
# cv2.imwrite("test.jpg", np.squeeze(x_crop).transpose((1,2,0)))
dboxes = detection_result["instances"].pred_boxes.tensor.cpu().detach().numpy()
dscores = detection_result["instances"].scores.cpu().detach().numpy()
center_pos = tracker.tracker.center_pos
# print(tracker.tracker.center_pos, tracker.tracker.size)
# print(dboxes, dscores)
# print(x_crop.shape)
all_outputs = tracker.track(frame, x_crop, scale_z, search_instance_size)
# cv2.imwrite("test.jpg", frame)
tboxes, tscores = all_outputs['bbox'], all_outputs['best_score']
# print(tboxes, tscores)
for idx, dbox in enumerate(dboxes):
required=False,
help="Path to the scikit-learn classifier")
ap.add_argument("-s",
"--scaler",
required=False,
help="Path to the scikit-learn scaler")
args = vars(ap.parse_args())
model_path: str = args.get("model", None)
if not model_path:
model_path = "./Model/svm.joblib"
model = joblib.load(model_path)
scaler_path: str = args.get("scaler", None)
if not scaler_path:
scaler_path = "./Model/svm_scaler.joblib"
scaler = joblib.load(scaler_path)
image = cv2.imread(args["image"])
image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
detect = Detection(model, scaler)
# Detected barcode image
barcode: np.ndarray = detect.detect(image,
window_size=(150, 250),
orientation=180,
pixels_per_cell=(150, 250),
cells_per_block=(1, 1),
threshold_proba=0.98,
threshold_overlap=0.15)
