python opencv流光

import cv2
import numpy as np

capture = cv2.VideoCapture('car.mp4')
feature_params = dict( maxCorners = 100,
     qualityLevel = 0.3,
     minDistance = 7,
    blockSize = 7 )

# lucas kanade光流参数
lk_params = dict( winSize = (15,15),
    maxLevel = 2,
    criteria = (cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10,
    0.03))

# 创建一些随机的颜色
color = np.random.randint(0,255,(100,3))
ret, old_frame = capture.read()
old_gray = cv2.cvtColor(old_frame,cv2.COLOR_BGR2GRAY)
p0 = cv2.goodFeaturesToTrack(old_gray,mask= None, **feature_params)

mask = np.zeros_like(old_frame)

while 1:
    ret,frame = capture.read()
    frame_gray = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)
    p1, st, err = cv2.calcOpticalFlowPyrLK(old_gray, frame_gray, p0, None,**lk_params)
    good_new = p1[st==1]
    good_old = p0[st==1]
    for i, (new, old) in enumerate(zip(good_new, good_old)):
        a, b = new.ravel()
        c, d = old.ravel()
        mask = cv2.line(mask, (a, b), (c, d), color[i].tolist(), 2)
        frame = cv2.circle(frame, (a, b), 5, color[i].tolist(), -1)
    img = cv2.add(frame, mask)
    cv2.imshow('frame', img)
    k = cv2.waitKey(30) & 0xff
    if k == 27:
        break
    old_gray = frame_gray.copy()
    p0 = good_new.reshape(-1, 1, 2)