def Change(self): for x in interface.user_interface.an.aliens: if common.distance(x.pos,self.pos)<.01: x.speed=abs(gauss(.5,.3)) x.Force=abs(gauss(.5,.3)) self._out() for x in interface.user_interface.an.predators: if common.distance(x.pos,self.pos)<.01: x.speed=abs(gauss(.5,.3)) x.Forse=abs(gauss(.5,.3)) self._out()
def findClosestStop(self):
return_dict = {}
for dir_tag in self.bus_dir_tag_dict:
#logging.info(dir_tag)
stop_loc_array = []
stop_id_array = []
for stop in self.stop_dict:
stop_loc_array.append([float(self.stop_dict[stop]['lat']),
float(self.stop_dict[stop]['lon'])])
stop_id_array.append(str(stop)) # TODO what is this really?
logging.info(self.stop_dict)
logging.info(stop_loc_array)
tree = KDTree(np.array(stop_loc_array))
for bus in self.bus_dir_tag_dict[dir_tag]:
if len(bus_id_array) > 1: #True:#try:
# TODO make this use bus heading to find the next stop, not the closest stop
logging.info("CLOSEST STOP TO BUS %s")
closest_bus_distance, index = tree.query(np.array([float(bus.lon), float(bus.lat)]),2)
for i in range(3):
closest_stop = stop_id_array[index[i]]
dist = common.distance(bus.lon, bus.lat,
stop_loc_array[index[i]][0], stop_loc_array[index[i]][1])
heading = common.heading_angle(bus.lon, bus.lat, stop_loc_array[index[i]][0], stop_loc_array[index[i]][1])
logging.info("STOP %s DIST %s HEADING %s" % (closest_stop, dist, heading))
bus.nextStopId = closest_stop
bus.distanceNextBus = float(dist)
return_dict[bus.vehicleId] = {'closest_stop' : closest_stop,
'closest_stop_distance' : dist}
else:#except:
return_dict[bus.vehicleId] = {'closest_stop' : None,
'closest_stop_distance' : None}
return return_dict
def findBunchingInRoute(self):
return_dict = {}
for dir_tag in self.bus_dir_tag_dict:
#logging.info(dir_tag)
bus_loc_array = []
bus_id_array = []
for bus in self.bus_dir_tag_dict[dir_tag]:
bus_loc_array.append([float(bus.lon), float(bus.lat)])
bus_id_array.append(str(bus.vehicleId)) # TODO what is this really?
#logging.info(bus_loc_array)
tree = KDTree(np.array(bus_loc_array))
for bus in self.bus_dir_tag_dict[dir_tag]:
if len(bus_id_array) > 1: #True:#try:
#logging.info(tree.query(np.array([float(bus.lon), float(bus.lat)]),2))
closest_bus_distance, index = tree.query(np.array([float(bus.lon), float(bus.lat)]),2)
closest_bus = bus_id_array[index[1]]
dist = common.distance(bus.lon, bus.lat,
bus_loc_array[index[1]][0], bus_loc_array[index[1]][1])
bus.nextBusId = closest_bus
bus.distanceNextBus = float(dist)
return_dict[bus.vehicleId] = {'closest_bus' : closest_bus,
'closest_bus_distance' : dist}
else:#except:
return_dict[bus.vehicleId] = {'closest_bus' : None,
'closest_bus_distance' : None}
return return_dict
def check_hit_by_bullet(self, bullets):
for enemy in self.enemies:
for bullet in bullets:
if distance(enemy.pos, bullet.pos) < 4:
enemy.hp -= bullet.damage
bullet.is_alive = False
self.score += 10 if enemy.hp <= 0 else 0
def check(self, fl):
if self.params["arg_alerts"] is False:
return
# log("Check Flight ", fl.data, fl.pos)
# Alert delayed waiting for position or least 16 messages
if fl.data['nm'] < 16:
if fl.pos['alt_ls'] == 0:
return None
if fl.pos['lat_ls'] == 0.0 and fl.pos['long_ls']:
return None
for alert in self.list:
# log("Check alert",alert.message())
if fl.data['ls'] - alert.fs < alert.alert[4]:
continue
if ((alert.alert[1] == 'ic' and alert.alert[2] == fl.data['ic']) or
(alert.alert[1] == 'sq' and alert.alert[2] == fl.data['sq']) or
(alert.alert[1] == 'cs' and alert.alert[2] == fl.data['cs'])):
alert.fs = fl.data['ls']
# log("Matching ", alert.message())
lat_ref = float(self.params["lat"])
long_ref = float(self.params["long"])
if fl.pos['lat_ls'] != 0.0 and fl.pos['long_ls'] != 0.0:
dist = distance(fl.pos['lat_ls'], fl.pos['long_ls'],
lat_ref, long_ref)
bear = bearing(lat_ref, long_ref, fl.pos['lat_ls'],
fl.pos['long_ls'])
else:
dist = bear = 0.0
alert.log(fl.data['ls'], fl.data['ic'], fl.pos['alt_ls'], dist,
bear)
return alert
return None
def calc_basis(cls, start_at):
reverse_dist = cls.calc_reverse(start_at)
old_dist = cls.calc_old_dist(from_node=start_at - 1)
roller = common.distance(cls.nodes[cls.solution[start_at - 1]],
cls.solution_nodes[start_at - 1 + 2:])
return reverse_dist, old_dist, roller
def print(self, _file=sys.stdout):
if _file == sys.stderr and (self.data['nm'] < 16 or
(time() - self.data['ls']) > 120):
return
lat_ref = float(self.params["lat"])
long_ref = float(self.params["long"])
dist = 0.0
bear = 0.0
if self.pos['lat_ls'] != 0.0 and self.pos['long_ls'] != 0.0:
dist = distance(self.pos['lat_ls'], self.pos['long_ls'], lat_ref,
long_ref)
bear = bearing(lat_ref, long_ref, self.pos['lat_ls'],
self.pos['long_ls'])
fmt = "{:s} " * 4 + "{:>8s} {:5d} {:7.1f}" + 4 * " {:5d}" + \
" {:9.5f} {:9.5f} {:5.1f} {:5.1f}° {:5.1f}°"
print(fmt.format(strftime("%d %H:%M:%S", gmtime(self.data['fs'])),
strftime("%d %H:%M:%S", gmtime(self.data['ls'])),
self.data['ic'], str(self.data['sq']),
str(self.data['cs']), self.data['nm'],
self.data['ls'] - self.data['fs'], self.pos['alt_fs'],
self.pos['alt_ls'], self.pos['alt_min'],
self.pos['alt_max'], self.pos['lat_ls'],
self.pos['long_ls'], dist, bear, self.pos['head_ls']),
file=_file)
def _evaluate_grad_attractive_potential(self, q: np.ndarray) -> np.ndarray:
d2goal = distance(q, self._goal)
if d2goal <= self._dstar:
grad_u_attr = self._zeta * (q - self._goal)
else:
grad_u_attr = self._dstar * self._zeta * (q - self._goal) / d2goal
return grad_u_attr
def _evaluate_attractive_potential(self, q: np.ndarray) -> float:
d2goal = distance(q, self._goal)
if d2goal <= self._dstar:
u_attr = .5 * self._zeta * d2goal**2
else:
u_attr = self._dstar * self._zeta * d2goal - .5 * self._zeta * self._dstar**2
return u_attr
def _move_to(self, i, j):
if self.intpos != None:
self.pos = deepcopy(interface.user_interface.setka[self.intpos[0]][self.intpos[1]])
self.intpos = None
self.movingto = [i, j]
if self.sj:
# значит х неизменный, идем по у
self.pos[1] += abs(self.timer.dt() * self.speed) * self.sj
elif self.si:
# значит y неизменный, идем по x
self.pos[0] += abs(self.timer.dt() * self.speed) * self.si
l = common.distance(self.pos, interface.user_interface.setka[self.movingto[0]][self.movingto[1]])
if l > self.prev:
self.intpos = deepcopy(self.movingto)
self.pos = deepcopy(interface.user_interface.setka[self.intpos[0]][self.intpos[1]])
self.movingto = None
self.prev = 100.0
else:
self.prev = common.distance(self.pos, interface.user_interface.setka[self.movingto[0]][self.movingto[1]])
def animsearch(self):
minim = [100500.0, 0, 0]
i = 0
for x in interface.user_interface.an.aliens:
m = common.distance(x.pos, self.pos)
if m < minim[0]:
minim[0] = m
minim[1] = i
i += 1
i = 0
if self.satiety < 0.1:
for x in interface.user_interface.an.predators:
m = common.distance(x.pos, self.pos)
if m < minim[0] and self.i != x.i:
minim[0] = m
minim[1] = i
minim[2] = 1
i += 1
if minim[0] == 100500.0:
return (0, 0)
if minim[2]:
if interface.user_interface.an.predators[minim[1]].intpos != None:
return (
interface.user_interface.an.predators[minim[1]].intpos[0] - self.intpos[0],
interface.user_interface.an.predators[minim[1]].intpos[1] - self.intpos[1],
)
else:
return (
interface.user_interface.an.predators[minim[1]].movingto[0] - self.intpos[0],
interface.user_interface.an.predators[minim[1]].movingto[1] - self.intpos[1],
)
else:
if interface.user_interface.an.aliens[minim[1]].intpos != None:
return (
interface.user_interface.an.aliens[minim[1]].intpos[0] - self.intpos[0],
interface.user_interface.an.aliens[minim[1]].intpos[1] - self.intpos[1],
)
else:
return (
interface.user_interface.an.aliens[minim[1]].movingto[0] - self.intpos[0],
interface.user_interface.an.aliens[minim[1]].movingto[1] - self.intpos[1],
)
def update(self):
Animal.update(self)
self.satiety -= 0.001
if self.satiety < 0.1:
for x in interface.user_interface.an.predators:
if common.distance(x.pos, self.pos) < 0.01 and self.i != x.i and self.alive:
self.fight(x)
for x in interface.user_interface.an.aliens:
if common.distance(x.pos, self.pos) < 0.01:
self.kill(x)
if self.intpos != None:
i, j = self.animsearch()
self.si = common.sign(i)
self.sj = common.sign(j)
self._move(i, j)
else:
self._move_to(self.movingto[0], self.movingto[1])
def calculate_path(self, start: np.ndarray, max_steps: int = 1000000, epsilon: float = None) -> np.ndarray:
epsilon = self._epsilon if epsilon is None else epsilon
path = np.zeros((max_steps, 2))
path[0] = start
step = 1
while distance(path[step - 1], self._goal) > EPS and step < max_steps:
grad = self._evaluate_grad_potential(path[step - 1])
path[step] = path[step - 1] - epsilon * grad
step += 1
return path[:step]
def _verify(
goal: np.ndarray, workspace: Circle, k: int,
default_epsilon: float) -> Tuple[np.ndarray, Circle, int, float]:
assert goal.size == 2, 'Goal must be a two-dimensional point'
r_goal = distance(goal, workspace.center)
assert r_goal <= workspace.radius, 'Goal must be inside the workspace'
assert k > 0, 'Parameter k must be a value greater than zero'
assert default_epsilon > 0, 'Epsilon must be a value greater than 0'
return goal.copy(), workspace, k, default_epsilon
def add_distance_to_dict(user_latitude, user_longitude, bus_stops_data):
""" This method modifies the bus stop list of dictionaries
and adds the distance of the bus stop from the user to each record.
"""
for data in bus_stops_data:
if "Distance" in data:
del data["Distance"] # this deletes the existing distance data
for data in bus_stops_data:
dist = distance(user_latitude, user_longitude, data["Latitude"],
data["Longitude"])
data[
"Distance"] = dist # add all the respective distance into the data in the list of data
return bus_stops_data
def calculate_path(self,
start: np.ndarray,
max_steps: int = 10000,
epsilon: float = None,
scale_magnitude: float = None) -> np.ndarray:
epsilon = self._epsilon if epsilon is None else epsilon
starting_distance = distance(start, self._goal)
path = np.zeros((max_steps, 2))
path[0] = start
step = 1
while distance(path[step - 1], self._goal) > EPS and step < max_steps:
grad = self._evaluate_grad_phi(path[step - 1])
if scale_magnitude is not None:
d2goal = min(distance(path[step - 1], self._goal),
starting_distance)
eta = (d2goal / starting_distance)**2
grad = eta * scale_magnitude * grad / np.linalg.norm(grad)
path[step] = path[step - 1] - epsilon * grad
step += 1
return path[:step]
def calc_connector(cls, roller, left, node_before_a, node_a):
to_d = roller[:-1].copy()
offset = left % 10
if node_before_a not in cls.prime_set and offset == 0:
to_d *= common.penalty
roller = common.distance(cls.nodes[node_a],
cls.solution_nodes[left + 2:])
if node_a in cls.prime_set:
return to_d, roller, roller
else:
start_leg = (10 - left - 2) % 10
from_a = roller.copy()
from_a[start_leg::10] *= common.penalty
return to_d, from_a, roller
def _evaluate_repulsive_potential(self, q: np.ndarray) -> float:
u_rep = 0
d2circle = self._workspace.radius - distance(q, self._workspace.center)
if d2circle < EPS:
u_rep += LARGE
elif abs(d2circle) <= self._workspace.safe_distance:
u_rep += .5 * self._eta * ((1 / d2circle) - (1 / self._workspace.safe_distance)) ** 2
for circle in self._obstacles:
d2circle = circle_distance(q, circle)
if d2circle < EPS:
u_rep += LARGE
elif d2circle <= circle.safe_distance:
u_rep += .5 * self._eta * ((1 / d2circle) - (1 / circle.safe_distance))**2
return u_rep
def test_bifurcation_angle(common_input, angle):
common_input.update(dict(keep_fixed_1 = False,
keep_fixed_2 = False,
bif = False,
lower = False,
cylinder_factor = 7,
angle = angle,
region_of_interest = "commandline",
region_points = [35.8, 59.8, 39.7, 76.8, 54.7, 53.2]))
rotate_branches(**common_input)
# Read in files to compute angle
base_path = get_path_names(common_input["input_filepath"])
old_centerlines = read_polydata(base_path + "_centerline_par.vtp")
new_centerlines = read_polydata(base_path + "_centerline_interpolated_ang.vtp")
end_points = read_polydata(base_path + "_clippingpoints.vtp")
# Start points
start_point1 = end_points.GetPoint(1)
start_point2 = end_points.GetPoint(2)
# Get relevant centerlines
cl_old_1 = -1
cl_old_2 = -1
cl_new_1 = -1
cl_new_2 = -1
tol = get_tolerance(old_centerlines)
for i in range(old_centerlines.GetNumberOfLines()):
line_old = extract_single_line(old_centerlines, i)
line_new = extract_single_line(new_centerlines, i)
loc_old = get_locator(line_old)
loc_new = get_locator(line_new)
id1_old = loc_old.FindClosestPoint(start_point1)
id2_old = loc_old.FindClosestPoint(start_point2)
id1_new = loc_new.FindClosestPoint(start_point1)
id2_new = loc_new.FindClosestPoint(start_point2)
if distance(start_point1, line_old.GetPoint(id1_old)) < tol:
cl_old_1 = i
cl_old_id1 = id1_old
if distance(start_point2, line_old.GetPoint(id2_old)) < tol:
cl_old_2 = i
cl_old_id2 = id2_old
if distance(start_point1, line_new.GetPoint(id1_new)) < tol:
cl_new_1 = i
cl_new_id1 = id1_new
if distance(start_point2, line_new.GetPoint(id2_new)) < tol:
cl_new_2 = i
cl_new_id2 = id2_new
if -1 not in [cl_old_1, cl_old_2, cl_new_1, cl_new_2]:
break
# Get end points
end_point1_old = np.array(extract_single_line(old_centerlines, cl_old_1).GetPoint(cl_old_id1 + 20))
end_point2_old = np.array(extract_single_line(old_centerlines, cl_old_2).GetPoint(cl_old_id2 + 20))
end_point1_new = np.array(extract_single_line(new_centerlines, cl_new_1).GetPoint(cl_new_id1 + 20))
end_point2_new = np.array(extract_single_line(new_centerlines, cl_new_2).GetPoint(cl_new_id2 + 20))
# Vectors
v1_old = end_point1_old - np.array(start_point1)
v2_old = end_point2_old - np.array(start_point2)
v1_new = end_point1_new - np.array(start_point1)
v2_new = end_point2_new - np.array(start_point2)
# Normalize
v1_old = v1_old / np.sqrt(np.sum(v1_old**2))
v2_old = v2_old / np.sqrt(np.sum(v2_old**2))
v1_new = v1_new / np.sqrt(np.sum(v1_new**2))
v2_new = v2_new / np.sqrt(np.sum(v2_new**2))
# Angle
first_daughter_branch_angle_change = np.arccos(np.dot(v1_old, v1_new))
second_daughter_branch_angle_change = np.arccos(np.dot(v2_old, v2_new))
assert abs(first_daughter_branch_angle_change
+ second_daughter_branch_angle_change
- 2 * abs(common_input["angle"])) < 0.01
def _evaluate_grad_gamma(self, q: np.ndarray) -> np.ndarray:
grad_gamma = 2 * self._k * distance(q, self._goal)**(
2 * self._k - 1) * (q - self._goal) / distance(q, self._goal)
return grad_gamma
def _evaluate_beta_i(q: np.ndarray, obstacle: Circle):
return distance(q, obstacle.center)**2 - obstacle.radius**2
def _evaluate_gamma(self, q: np.ndarray) -> float:
gamma = distance(q, self._goal)**(2 * self._k)
return gamma
def bar_code(binary, gray, min_ratio=2, min_area=80, max_area=30000):
# 中值滤波
img = cv2.medianBlur(binary, 3)
# 形态学处理
img = common.dilate_(img, ksize=(3, 3), iterations=4)
_, contours, hierarchy = cv2.findContours(img, cv2.RETR_LIST,
cv2.CHAIN_APPROX_SIMPLE)
def max_filter(tour):
area = cv2.contourArea(tour)
if min_area <= area <= max_area:
return True
return False
contours = filter(max_filter, contours)
contours = sorted(contours, key=cv2.contourArea, reverse=True)
contours = contours[:min(25, int(0.2 * len(contours)))]
angles = {}
for cnt in contours:
rect = cv2.minAreaRect(cnt)
box = np.int0(cv2.boxPoints(rect))
if min_ratio <= common.distance(box[0], box[1]) / common.distance(
box[1], box[2]):
if box[1][0] - box[0][0] == 0:
angle = 90
else:
angle = math.atan((box[1][1] - box[0][1]) /
(box[1][0] - box[0][0])) * 180 / np.pi
elif min_ratio <= common.distance(box[1], box[2]) / common.distance(
box[0], box[1]):
if box[1][0] - box[2][0] == 0:
angle = 90
else:
angle = math.atan((box[1][1] - box[2][1]) /
(box[1][0] - box[2][0])) * 180 / np.pi
else:
continue
# cv2.drawContours(gray, [box], -1, (200, 200, 200), 3)
key = (angle + 75) / 30
if key < 0:
key = key + 6
angle = angle + 180
key = int(key)
value = angles.setdefault(key, [])
value.append(angle)
angles = sorted(angles.items(),
key=lambda item: len(item[1]),
reverse=True)
key, value = angles[0]
rot_angle = np.median(value)
h, w = gray.shape[:2]
if abs(rot_angle) > 45:
side = max(w, h)
center = (side / 2, side / 2)
new_shape = (side, side)
else:
center = (w / 2, h / 2)
new_shape = (w, h)
M = cv2.getRotationMatrix2D(center, rot_angle, 1)
gray = cv2.warpAffine(gray, M, new_shape)
return gray
def check_hit_by_enemy(self, enemies):
for enemy in enemies:
if distance(self.player.pos, enemy.pos) < 4 and not self.player.invincible_time > 0:
self.player.life -= 1
self.player.invincible_time = PLAYER_INVINCIBLE_TIME
def check_hit_by_bullet(self, bullets):
for bullet in bullets:
if distance(self.player.pos, bullet.pos) < 4 and not self.player.invincible_time > 0:
self.player.life -= 1
self.player.invincible_time = PLAYER_INVINCIBLE_TIME
bullet.is_alive = False
