光流法学习笔记

参考博客

光流法简介
https://blog.csdn.net/qq_41368247/article/details/82562165

基于金字塔分层的LK光流
https://blog.csdn.net/sgfmby1994/article/details/68489944

python代码

LK算法（视频）

import numpy as np
import cv2
cap = cv2.VideoCapture('./data/mv1.mp4')
# params for ShiTomasi corner detection
feature_params = dict( maxCorners = 100,
qualityLevel = 0.3,
minDistance = 7,
blockSize = 7 )
# Parameters for lucas kanade optical flow
#maxLevel 为使用的图像金字塔层数
lk_params = dict( winSize = (15,15),
maxLevel = 2,
criteria = (cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10, 0.03))
# Create some random colors
color = np.random.randint(0,255,(100,3))
# Take first frame and find corners in it
ret, old_frame = cap.read()
old_gray = cv2.cvtColor(old_frame, cv2.COLOR_BGR2GRAY)
p0 = cv2.goodFeaturesToTrack(old_gray, mask = None, **feature_params)
# Create a mask image for drawing purposes
mask = np.zeros_like(old_frame)
cv2.namedWindow('frame', cv2.WINDOW_NORMAL)
while(1):
    ret,frame = cap.read()
    frame_gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    # calculate optical flow 能够获取点的新位置
    p1, st, err = cv2.calcOpticalFlowPyrLK(old_gray, frame_gray, p0, None, **lk_params)
    # Select good points
    good_new = p1[st==1]
    good_old = p0[st==1]
    # draw the tracks
    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
    # Now update the previous frame and previous points
    old_gray = frame_gray.copy()
    p0 = good_new.reshape(-1,1,2)
cv2.destroyAllWindows()
cap.release()

LK算法（前后帧）

#!/usr/bin/env python 
# -*- coding:utf-8 -*-
import cv2
import numpy as np


prev = cv2.imread('./data/prev.jpg')
now = cv2.imread('./data/now.jpg')

# 角点检测的参数
feature_params = dict(maxCorners=100,
                      qualityLevel=0.3,
                      minDistance=7,
                      blockSize=7)

# 光流法参数
# maxLevel 未使用的图像金字塔层数（金字塔在前面说了是为了解决速度过快）
lk_params = dict(winSize=(15, 15),
                 maxLevel=2,
                 criteria=(cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10, 0.03))

# 创建随机生成的颜色
color = np.random.randint(0, 255, (100, 3))
color = np.int0(color)

# 获得角点
prev_gray = cv2.cvtColor(prev, cv2.COLOR_BGR2GRAY)
p0 = cv2.goodFeaturesToTrack(prev_gray, mask=None, **feature_params)
# print(p0)
cornors = np.int0(p0)  # 将浮点数转化为整数

# 创建掩码
mask = np.zeros_like(prev)
print(mask)
# 绘制角点
# i = 0
# for cornor in cornors:
#     x,y = cornor.ravel()
#     cv2.circle(prev,(x,y),1,color[i].tolist(),10)
#     i += 1

# 将后面的图片也转化为灰度图
now_gray = cv2.cvtColor(now, cv2.COLOR_BGR2GRAY)
# cv2.imshow('now', now_gray)

# 通过光流算法计算出之前的角点下一帧的位置
p1, st, err = cv2.calcOpticalFlowPyrLK(prev_gray, now_gray, p0, None, **lk_params)


# 选择有效的角点坐标
good_new = p1[st==1]
good_old = p0[st==1]


# 绘制光流
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(now,(a,b),5,color[i].tolist(),-1)
    img = cv2.add(frame,mask)


cv2.imshow('frame', prev)
cv2.waitKey(200)
# cv2.imshow('now', now)
cv2.imshow('frame',img)

# prev_gray[p0] == [255,255,255]
# cv2.imshow('prev',prev)
cv2.waitKey(0)

稠密光流（视频）

import cv2
import numpy as np
cap = cv2.VideoCapture("./data/mv1.mp4")
ret, frame1 = cap.read()
prvs = cv2.cvtColor(frame1,cv2.COLOR_BGR2GRAY)
hsv = np.zeros_like(frame1)
hsv[...,1] = 255
while(1):
    ret, frame2 = cap.read()
    next = cv2.cvtColor(frame2,cv2.COLOR_BGR2GRAY)

    flow = cv2.calcOpticalFlowFarneback(prvs,next, None, 0.5, 3, 15, 3, 5, 1.2, 0)
    #将直角坐标转化为极坐标
    mag, ang = cv2.cartToPolar(flow[...,0], flow[...,1])
    hsv[...,0] = ang*180/np.pi/2
    hsv[...,2] = cv2.normalize(mag,None,0,255,cv2.NORM_MINMAX)
    rgb = cv2.cvtColor(hsv,cv2.COLOR_HSV2BGR)
    cv2.imshow('frame2',rgb)
    k = cv2.waitKey(30) & 0xff
    if k == 27:
        break
    elif k == ord('s'):
        cv2.imwrite('opticalfb.png',frame2)
        cv2.imwrite('opticalhsv.png',rgb)
    prvs = next

cap.release()
cv2.destroyAllWindows()

稠密光流（前后帧）

#!/usr/bin/env python 
# -*- coding:utf-8 -*-
import cv2
import numpy as np
prev = cv2.imread('./data/prev.jpg')
now = cv2.imread('./data/now.jpg')
prev_gray = cv2.cvtColor(prev,cv2.COLOR_BGR2GRAY)
hsv = np.zeros_like(prev)
hsv[...,1] = 255
# print(hsv)
# cv2.imshow('frame', hsv)
# cv2.waitKey(0)

now_gray = cv2.cvtColor(now, cv2.COLOR_BGR2GRAY)
flow = cv2.calcOpticalFlowFarneback(prev_gray,now_gray, None, 0.5, 3, 15, 3, 5, 1.2, 0)
mag, ang = cv2.cartToPolar(flow[...,0], flow[...,1])
hsv[...,0] = ang*180/np.pi/2
hsv[...,2] = cv2.normalize(mag,None,0,255,cv2.NORM_MINMAX)
rgb = cv2.cvtColor(hsv,cv2.COLOR_HSV2BGR)
cv2.imshow('frame2',rgb)
cv2.waitKey(0)