本文整理汇总了Python中pykalman.KalmanFilter.transition_covariance方法的典型用法代码示例。如果您正苦于以下问题：Python KalmanFilter.transition_covariance方法的具体用法？Python KalmanFilter.transition_covariance怎么用？Python KalmanFilter.transition_covariance使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类pykalman.KalmanFilter的用法示例。
在下文中一共展示了KalmanFilter.transition_covariance方法的2个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: demo
# 需要导入模块: from pykalman import KalmanFilter [as 别名]
# 或者: from pykalman.KalmanFilter import transition_covariance [as 别名]
def demo():
img = np.zeros((768, 1024, 3), np.uint8)
trajs = deque(maxlen=200)
trajs_filterd = deque(maxlen=200)
def draw(event, x, y, flags, param):
kf, fm, fo = param
if event == cv2.EVENT_MOUSEMOVE:
fm[:], fo[:] = kf.filter_update(fm, fo, np.array([x, y]))
fx = int(fm[0])
fy = int(fm[1])
trajs.append((x, y))
trajs_filterd.append((fx, fy))
cv2.namedWindow("image")
kf = KalmanFilter(n_dim_state=4, n_dim_obs=2)
kf.transition_matrices = np.array([[1, 0, 1, 0], [0, 1, 0, 1], [0, 0, 1, 0], [0, 0, 0, 1]])
kf.observation_matrices = np.array([[1, 0, 0, 0], [0, 1, 0, 0]])
kf.transition_covariance = 1e-3 * np.eye(4)
kf.observation_covariance = 10 * np.eye(2)
kf.initial_state_mean = np.array([0, 0, 0, 0])
kf.initial_state_covariance = np.eye(4)
fm = kf.initial_state_mean
fo = kf.initial_state_covariance
cv2.setMouseCallback("image", draw, (kf, fm, fo))
while True:
img = np.zeros((768, 1024, 3), np.uint8)
for i in range(len(trajs) - 1):
cv2.line(img, trajs[i], trajs[i + 1], (0, 255, 0), 1)
cv2.line(img, trajs_filterd[i], trajs_filterd[i + 1], (0, 0, 255), 1)
map(lambda pt: cv2.circle(img, pt, 3, (0, 255, 0), 1), trajs)
map(lambda pt: cv2.circle(img, pt, 3, (0, 0, 255), 1), trajs_filterd)
cv2.imshow("image", img)
if cv2.waitKey(5) & 0xFF == ord("q"):
return示例2: range
# 需要导入模块: from pykalman import KalmanFilter [as 别名]
# 或者: from pykalman.KalmanFilter import transition_covariance [as 别名]
import random
N = 3 * 500
sigma = 0.7
# Function to estimate
W = numpy.zeros(N)
W[0 : 500 ] = 1
W[500 : 1000 ] = 3
W[1000 : 1500 ] = 2
# Generate noisy measures
X = numpy.zeros(N)
for i in range(N):
X[i] = W[i] + random.gauss(0, sigma)
kf = KalmanFilter(initial_state_mean=X[0], n_dim_obs=1)
kf.transition_covariance = 1e-3
Z = kf.em(X).smooth(X)[0]
#Z = kf.smooth(X)[0]
plt.figure(1)
plt.plot( X, 'r:' )
plt.plot( W, 'b-' )
plt.plot( Z, 'g-' )
plt.show()