opencv-python DIS光流

Posted -牧野-

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opencv-python DIS光流相关的知识,希望对你有一定的参考价值。

# -牧野- 2021-05-29
import cv2
from flowlib import flow_to_image
import numpy as np

def warp_flow(img, flow):
    h, w = flow.shape[:2]
    flow = -flow
    flow[:,:,0] += np.arange(w)
    flow[:,:,1] += np.arange(h)[:,np.newaxis]
    res = cv2.remap(img, flow, None, cv2.INTER_LINEAR)
    return res

img0_path = 'im1.png'
img1_path = 'im3.png'

img0 = cv2.imread(img0_path)
img1 = cv2.imread(img1_path)

img0_gray = cv2.cvtColor(img0, cv2.COLOR_BGR2GRAY)
img1_gray = cv2.cvtColor(img1, cv2.COLOR_BGR2GRAY)

dis = cv2.DISOpticalFlow_create(2)  #PRESET_ULTRAFAST, PRESET_FAST and PRESET_MEDIUM
flow0 = dis.calc(img0_gray, img1_gray, None, )
flow1 = dis.calc(img1_gray, img0_gray, None, )

img1_warp = warp_flow(img0, flow0)
img0_warp = warp_flow(img1, flow1)

flow0_img = flow_to_image(flow0)
flow1_img = flow_to_image(flow1)

cv2.imwrite('flow0.jpg', flow0_img)
cv2.imwrite('flow1.jpg', flow1_img)
cv2.imwrite('im3_warp.jpg', img1_warp)
cv2.imwrite('im1_warp.jpg', img0_warp)

引用的flowlib.py文件见https://github.com/liruoteng/OpticalFlowToolkit/blob/master/lib/flowlib.py

输入图像:

 

0:PRESET_ULTRAFAST 1:PRESET_FAST   2:PRESET_MEDIUM对应结果:

warp结果:

以上是关于opencv-python DIS光流的主要内容,如果未能解决你的问题,请参考以下文章

opencv-python DIS光流

opencv-python DIS光流

OpenCV4中DIS光流算法与应用

OpenCV-Python视频分析(移动物体检测,物体追踪)

有没有光流代码的集合?

基于 OpenCV C++ 光流的分割代码抛出异常