def drawNetworkWithHomography(self, H):
edges = self.roadNetwork.edges
nodes = self.roadNetwork.nodes
for edge in edges:
nodeI = edge.node1
nodeJ = edge.node2
width = edge.width
### Prastutaniki angle em ledhu, only straight lines.
pts1 = np.array([[int(nodeI.x-width/2), nodeI.y], [int(nodeI.x-width/2), nodeJ.y]])
pts2 = np.array([[int(nodeI.x+width/2), nodeI.y], [int(nodeI.x+width/2), nodeJ.y]])
pts1 = transformSetOfPointsAndReturn(pts1, H)
pts2 = transformSetOfPointsAndReturn(pts2, H)
cv2.line(self.grid, (pts1[0][0], pts1[0][1]), (pts1[1][0], pts1[1][1]), (255,0,0), 1)
cv2.line(self.grid, (pts2[0][0], pts2[0][1]), (pts2[1][0], pts2[1][1]), (255,0,0), 1)
def simulateFromXML(video, xml, size, homography=False):
frames , count = getAllFrames(xml)
if video != None:
cap = cv2.VideoCapture(video)
for i, frame in enumerate(frames.keys()):
vehicles = frames[frame]
print "Frame Number ", i
if video != None:
ret, video_frame = cap.read()
#print frames[frame]
output = np.zeros(size)
output_from_model = np.zeros(size)
#cv2.putText(output, str(i), (20,20), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (255,0,0), 1, cv2.CV_AA)
if homography == True:
straight_road_points = np.array([[0,0],[300,0],[300,710],[0,710]], dtype=float)
perspective_road_points = np.array([[536,136], [615,136], [655,655], [0,475]], dtype=float)
perspective_road_points1 = np.array([[536,136], [615,136], [655,655], [0,475], [ 414., 464], [549., 245]], dtype=float)
straight_road_points1 = np.array([[0,0],[300,0],[300,710],[0,710], [150,0], [150,100] , [150,200], [150,300], [150, 400], [150, 500], [150, 600], [150, 700]], dtype=float)
output = drawPoly(output, straight_road_points, [255,0,0])
output = drawPoly(output, perspective_road_points, [0,0,255])
video_frame = drawPoly(video_frame, straight_road_points, [255,0,0])
video_frame = drawPoly(video_frame, perspective_road_points, [0,0,255])
H, status = cv2.findHomography(perspective_road_points, straight_road_points)
#H, status = cv2.findHomography(straight_road_points, perspective_road_points)
straight_road_points2 = np.copy(straight_road_points1)
#print perspective_road_points
#for (x, y) in zip(straight_road_points2, transformSetOfPointsAndReturn(straight_road_points1 ,H)):
# print x, y
#break
vehicles_topView = np.copy(vehicles)
vehicles_topView = changeVehicles(transformSetOfPointsAndReturn(getVehicleCoordinates(vehicles), H), vehicles_topView)
print vehicles, transformSetOfPointsAndReturn(getVehicleCoordinates(vehicles), H), vehicles_topView
output = drawVehicles(output, vehicles)
if homography == True:
output = drawVehicles_red(output, vehicles_topView)
output = output[::-1,:,:]
#output = output[:,::-1,:]
cv2.imshow("sim", output)
if video != None:
cv2.imshow("real", video_frame)
if cv2.waitKey(33) == 27:
break
cv2.destroyAllWindows()
def simulateFromXML(video, xml, size, homography=False):
frames, count = getAllFrames(xml)
if video != None:
cap = cv2.VideoCapture(video)
prev_vehicles = np.array([])
simulated_vehicles = None
accelerating_vehicles = {}
left_vehicles = []
prev_speeds = {}
for i, frame in enumerate(frames.keys()):
vehicles = frames[frame]
print "Frame Number ", i, len(vehicles)
if video != None:
ret, video_frame = cap.read()
#print frames[frame]
output = np.zeros(size)
cv2.putText(output, str(i), (20, 20), cv2.FONT_HERSHEY_SIMPLEX, 0.75,
(255, 0, 0), 1, cv2.CV_AA)
output_from_model = np.zeros(size)
if homography == True:
straight_road_points = np.array(
[[0, 0], [300, 0], [300, 710], [0, 710]], dtype=float)
perspective_road_points = np.array(
[[536, 136], [615, 136], [655, 655], [0, 475]], dtype=float)
perspective_road_points1 = np.array(
[[536, 136], [615, 136], [655, 655], [0, 475], [414., 464],
[549., 245]],
dtype=float)
straight_road_points1 = np.array(
[[0, 0], [300, 0], [300, 710], [0, 710], [150, 0], [150, 100],
[150, 200], [150, 300], [150, 400], [150, 500], [150, 600],
[150, 700]],
dtype=float)
output = drawPoly(output, straight_road_points, [255, 0, 0])
output = drawPoly(output, perspective_road_points, [0, 0, 255])
video_frame = drawPoly(video_frame, straight_road_points,
[255, 0, 0])
video_frame = drawPoly(video_frame, perspective_road_points,
[0, 0, 255])
H, status = cv2.findHomography(perspective_road_points,
straight_road_points)
#H, status = cv2.findHomography(straight_road_points, perspective_road_points)
straight_road_points2 = np.copy(straight_road_points1)
#print perspective_road_points
#for (x, y) in zip(straight_road_points2, transformSetOfPointsAndReturn(straight_road_points1 ,H)):
# print x, y
#break
vehicles_topView = np.copy(vehicles)
vehicles_topView = changeVehicles(
transformSetOfPointsAndReturn(getVehicleCoordinates(vehicles),
H), vehicles_topView)
print vehicles, transformSetOfPointsAndReturn(
getVehicleCoordinates(vehicles), H), vehicles_topView
#output = drawVehicles(output, vehicles)
for vehicle in vehicles:
for veh in prev_vehicles:
if vehicle[0] == veh[0]:
print "difference ", veh, vehicle, abs(veh[4] - vehicle[4])
if abs(veh[4] - vehicle[4]) > 2:
if veh[0] not in accelerating_vehicles:
accelerating_vehicles[veh[0]] = 1
else:
accelerating_vehicles[veh[0]] += 1
#vehicle[5] = vehicle[4] - veh[4]
#vehicle[5] = 0
else:
vehicle[5] = 0
if i == 0:
simulated_vehicles = np.copy(vehicles)
else:
simulated_vehicles1 = np.copy(vehicles)
print "Simulated Vehicles 1 ", simulated_vehicles
vehicles_inframe = []
for j, veh in enumerate(simulated_vehicles1):
for v in simulated_vehicles:
if v[0] == veh[0] and (v[4] != 0):
#veh[1] = v[1]
if v[4] + veh[5] <= 17:
#if v[4] + veh[5] > -1:
veh[2] = v[2] + veh[5]
veh[4] = v[4] + veh[5]
#veh[5] = v[5]
else:
veh[2] = v[2]
veh[4] = 16
if veh[2] <= 680 and veh[0] not in left_vehicles:
vehicles_inframe.append(j)
elif veh[2] > 680:
left_vehicles.append(veh[0])
print "inframe ", vehicles_inframe, len(
vehicles_inframe), simulated_vehicles1.shape
print "before ", simulated_vehicles1
if simulated_vehicles1.size != 0:
simulated_vehicles1 = simulated_vehicles1[vehicles_inframe, :]
print "after ", simulated_vehicles1
simulated_vehicles = simulated_vehicles1
output, simulated_vehicles, prev_speeds = drawVehiclesWithSimulation(
output, vehicles, prev_vehicles, simulated_vehicles, prev_speeds)
prev_vehicles = np.copy(vehicles)
#output = drawVehicles(output, vehicles, 0)
if homography == True:
output = drawVehicles_red(output, vehicles_topView)
output = output[::-1, :, :]
#output = output[:,::-1,:]
cv2.imshow("sim", output)
print "Number of vehicles accelerating ", len(
accelerating_vehicles.keys())
if video != None:
cv2.imshow("real", video_frame)
print "Accelerating vehicles ", accelerating_vehicles
print "Number is ", len(accelerating_vehicles.keys())
if cv2.waitKey(33) == 27:
break
cv2.destroyAllWindows()