光流法draw_flow()函数报错

Posted jqpy1994

tags:

篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了光流法draw_flow()函数报错相关的知识,希望对你有一定的参考价值。

 光流法draw_flow()函数报错

 1 import cv2
 2 from scipy import *
 3 
 4 
 5 def draw_flow(im, flow, step=16):
 6     """ Plot optical flow at sample points
 7     spaced step pixels apart. """
 8     h, w = im.shape[:2]
 9     y, x = mgrid[step / 2:h:step, step / 2:w:step].reshape(2, -1)
10     fx,fy = flow[y, x].T
11     # create line endpoints
12     lines = vstack([x, y, x + fx, y + fy]).T.reshape(-1, 2, 2)
13     lines = int32(lines)
14     # create image and draw
15     vis = cv2.cvtColor(im, cv2.COLOR_GRAY2BGR)
16     for (x1, y1), (x2, y2) in lines:
17         cv2.line(vis, (x1, y1), (x2, y2), (0, 255, 0), 1)
18     cv2.circle(vis, (x1, y1), 1, (0, 255, 0), -1)
19     return vis
20 # setup video capture
21 cap = cv2.VideoCapture(0)
22 ret, im = cap.read()
23 prev_gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
24 while True:
25 # get grayscale image
26     ret, im = cap.read()
27     gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
28     # compute flow
29     flow = cv2.calcOpticalFlowFarneback(prev_gray, gray, None, 0.5, 3, 15, 3, 5, 1.2, 0)
30     prev_gray = gray
31     # plot the flow vectors
32     cv2.imshow(Optical flow, draw_flow(gray, flow))
33     if cv2.waitKey(10) == 27:
34         break

 

报错:

技术图片

解决方案行加上.astype(int)就解决了

y, x = mgrid[step / 2:h:step, step / 2:w:step].reshape(2, -1).astype(int) #以网格的形式选取二维图像上等间隔的点,这里间隔为16,reshape成2行的array

以上是关于光流法draw_flow()函数报错的主要内容,如果未能解决你的问题,请参考以下文章

OpenCV LK光流法测试

光流法详解之一(LK光流)

目标跟踪之Horn-Schunck光流法

光流法

车辆计数基于matlab光流法行驶车辆检测计数含Matlab源码 627期

光流法详解之二(HS光流)