# 光流追蹤
# 光流追蹤的前提是：1. 對(duì)象的像素強(qiáng)度在連續(xù)幀之間不會(huì)改變；2. 相鄰像素具有相似的運(yùn)動(dòng)。
# - cv2.goodFeaturesToTrack() 確定要追蹤的特征點(diǎn)
# - cv2.calcOpticalFlowPyrLK() 追蹤視頻中的特征點(diǎn)
# 取第一幀，檢測(cè)其中的一些 Shi-Tomasi 角點(diǎn)，使用 Lucas-Kanade 光流迭代跟蹤這些點(diǎn)。
# 對(duì)于函數(shù) cv2.calcOpticalFlowPyrLK() 傳遞前一幀、前一個(gè)點(diǎn)和下一幀。它返回下一個(gè)點(diǎn)以及一些狀態(tài)編號(hào)，如果找到下一個(gè)點(diǎn)，則值為 1，否則為零。
# 然后在下一步中迭代地將這些下一個(gè)點(diǎn)作為前一個(gè)點(diǎn)傳遞。
# USAGE
# python video_optical_flow.py
import imutils
import numpy as np
import cv2
cap = cv2.VideoCapture('images/slow_traffic_small.mp4')
# ShiTomasi角點(diǎn)檢測(cè)的參數(shù)
feature_params = dict(maxCorners=100,
 qualityLevel=0.3,
 minDistance=7,
 blockSize=7)
# Lucas Kanada光流檢測(cè)的參數(shù)
lk_params = dict(winSize=(15, 15),
  maxLevel=2,
  criteria=(cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10, 0.03))
# 構(gòu)建隨機(jī)顏色
color = np.random.randint(0, 255, (100, 3))
# 獲取第一幀并發(fā)現(xiàn)角點(diǎn)
ret, old_frame = cap.read()
old_gray = cv2.cvtColor(old_frame, cv2.COLOR_BGR2GRAY)
p0 = cv2.goodFeaturesToTrack(old_gray, mask=None, **feature_params)
# 為繪制光流追蹤圖，構(gòu)建一個(gè)Mask
mask = np.zeros_like(old_frame)
num = 0
while (1):
 ret, frame = cap.read()
 if not ret:
  break
 frame_gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
 # 使用迭代Lucas Kanade方法計(jì)算稀疏特征集的光流
 # - old_gray: 上一幀單通道灰度圖
 # - frame_gray: 下一幀單通道灰度圖
 # - prePts：p0上一幀坐標(biāo)pts
 # - nextPts: None
 # - winSize: 每個(gè)金字塔級(jí)別上搜索窗口的大小
 # - maxLevel: 最大金字塔層數(shù)
 # - criteria：指定迭代搜索算法的終止條件，在指定的最大迭代次數(shù)criteria.maxCount之后或搜索窗口移動(dòng)小于criteria.epsilon
 p1, st, err = cv2.calcOpticalFlowPyrLK(old_gray, frame_gray, p0, None, **lk_params)
 # 選擇軌跡點(diǎn)
 good_new = p1[st == 1]
 good_old = p0[st == 1]
 # 繪制軌跡
 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)
 cv2.imwrite('videoof-imgs/' + str(num) + '.jpg', imutils.resize(img, 500))
 print(str(num))
 num = num + 1
 k = cv2.waitKey(30) & 0xff
 if k == 27:
  break
 # 更新之前的幀和點(diǎn)
 old_gray = frame_gray.copy()
 p0 = good_new.reshape(-1, 1, 2)
cv2.destroyAllWindows()
cap.release()

# 優(yōu)化后的光流追蹤—Lucas-Kanade tracker
# （當(dāng)不見(jiàn)檢查下一個(gè)關(guān)鍵點(diǎn)的正確程度時(shí)，即使圖像中的任何特征點(diǎn)消失，光流也有可能找到下一個(gè)看起來(lái)可能靠近它的點(diǎn)。實(shí)際上對(duì)于穩(wěn)健的跟蹤，角點(diǎn)應(yīng)該在特定的時(shí)間間隔內(nèi)檢測(cè)點(diǎn)。
# 找到特征點(diǎn)后，每 30 幀對(duì)光流點(diǎn)的向后檢查，只選擇好的。）
# Lucas Kanade稀疏光流演示。使用GoodFeatures跟蹤用于跟蹤初始化和匹配驗(yàn)證的回溯幀之間。
# Lucas-Kanade sparse optical flow demo. Uses goodFeaturesToTrack for track initialization and back-tracking for match verification between frames.
# Usage
# pyhton lk_track.py images/slow_traffic_small.mp4
# 按 ESC鍵退出
from __future__ import print_function
import imutils
import numpy as np
import cv2

def draw_str(dst, target, s):
 x, y = target
 cv2.putText(dst, s, (x + 1, y + 1), cv2.FONT_HERSHEY_PLAIN, 1.0, (0, 0, 0), thickness=2, lineType=cv2.LINE_AA)
 cv2.putText(dst, s, (x, y), cv2.FONT_HERSHEY_PLAIN, 1.0, (255, 255, 255), lineType=cv2.LINE_AA)

lk_params = dict(winSize=(15, 15),
  maxLevel=2,
  criteria=(cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10, 0.03))
feature_params = dict(maxCorners=500,
 qualityLevel=0.3,
 minDistance=7,
 blockSize=7)

class App:
 def __init__(self, video_src):
  self.track_len = 10
  self.detect_interval = 30
  self.tracks = []
  self.cam = cv2.VideoCapture(video_src)
  self.frame_idx = 0
 def run(self):
  while True:
_ret, frame = self.cam.read()
if not _ret:
 break
frame_gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
vis = frame.copy()
if len(self.tracks) > 0:
 img0, img1 = self.prev_gray, frame_gray
 p0 = np.float32([tr[-1] for tr in self.tracks]).reshape(-1, 1, 2)
 p1, _st, _err = cv2.calcOpticalFlowPyrLK(img0, img1, p0, None, **lk_params)
 p0r, _st, _err = cv2.calcOpticalFlowPyrLK(img1, img0, p1, None, **lk_params)
 d = abs(p0 - p0r).reshape(-1, 2).max(-1)
 good = d < 1
 new_tracks = []
 for tr, (x, y), good_flag in zip(self.tracks, p1.reshape(-1, 2), good):
  if not good_flag:continue
  tr.append((x, y))
  if len(tr) > self.track_len:del tr[0]
  new_tracks.append(tr)
  cv2.circle(vis, (x, y), 2, (0, 255, 0), -1)
 self.tracks = new_tracks
 cv2.polylines(vis, [np.int32(tr) for tr in self.tracks], False, (0, 255, 0))
 draw_str(vis, (20, 20), 'track count: %d' % len(self.tracks))
if self.frame_idx % self.detect_interval == 0:
 mask = np.zeros_like(frame_gray)
 mask[:] = 255
 for x, y in [np.int32(tr[-1]) for tr in self.tracks]:
  cv2.circle(mask, (x, y), 5, 0, -1)
 p = cv2.goodFeaturesToTrack(frame_gray, mask=mask, **feature_params)
 if p is not None:
  for x, y in np.float32(p).reshape(-1, 2):self.tracks.append([(x, y)])
self.prev_gray = frame_gray
cv2.imshow('lk_track', vis)
print(self.frame_idx)
cv2.imwrite('videoOof-imgs/' + str(self.frame_idx) + '.jpg', imutils.resize(vis, 500))
self.frame_idx += 1
ch = cv2.waitKey(1)
if ch == 27:
 break
def main():
 import sys
 try:
  video_src = sys.argv[1]
 except:
  video_src = 0
 App(video_src).run()
 print('Done')

if __name__ == '__main__':
 print(__doc__)
 main()
 cv2.destroyAllWindows()

# OpenCV中的密集光流
# Lucas-Kanade 方法計(jì)算稀疏特征集的光流（使用 Shi-Tomasi 算法檢測(cè)到的角點(diǎn)）。
# OpenCV 提供了另一種算法: Gunner Farneback 來(lái)尋找密集光流。它計(jì)算幀中所有點(diǎn)的光流。
# 通過(guò)cv2.calcOpticalFlowFarneback() 將得到一個(gè)帶有光流向量 (u,v) 的 2 通道陣列?？梢哉业剿鼈兊拇笮『头较?，然后對(duì)結(jié)果進(jìn)行顏色編碼以實(shí)現(xiàn)更好的可視化。
# 在HSV圖像中，方向?qū)?yīng)于圖像的色調(diào)，幅度對(duì)應(yīng)于價(jià)值平面。
import cv2
import imutils
import numpy as np
cap = cv2.VideoCapture('images/slow_traffic_small.mp4')
ret, frame1 = cap.read()
prvs = cv2.cvtColor(frame1, cv2.COLOR_BGR2GRAY)
hsv = np.zeros_like(frame1)
hsv[..., 1] = 255
num = 0
while (1):
 ret, frame2 = cap.read()
 if not ret:
  break
 next = cv2.cvtColor(frame2, cv2.COLOR_BGR2GRAY)
 # 使用迭代Gunner Farneback 方法計(jì)算密集特征的光流
 # - prvs: 上一幀單通道灰度圖
 # - next: 下一幀單通道灰度圖
 # - flow: 流 None
 # - pyr_scale: 0.5經(jīng)典金字塔，構(gòu)建金字塔縮放scale
 # - level：3 初始圖像的金字塔層數(shù)
 # - winsize：3 平均窗口大小，數(shù)值越大，算法對(duì)圖像的魯棒性越強(qiáng)
 # - iterations：15 迭代次數(shù)
 # - poly_n：5 像素鄰域的參數(shù)多邊形大小，用于在每個(gè)像素中找到多項(xiàng)式展開(kāi)式；較大的值意味著圖像將使用更平滑的曲面進(jìn)行近似，從而產(chǎn)生更高的分辨率、魯棒算法和更模糊的運(yùn)動(dòng)場(chǎng)；通常多邊形n=5或7。
 # - poly_sigma：1.2 高斯標(biāo)準(zhǔn)差，用于平滑導(dǎo)數(shù)
 # - flags: 可以是以下操作標(biāo)志的組合：OPTFLOW_USE_INITIAL_FLOW：使用輸入流作為初始流近似值。OPTFLOW_FARNEBACK_GAUSSIAN: 使用GAUSSIAN過(guò)濾器而不是相同尺寸的盒過(guò)濾器；
 flow = cv2.calcOpticalFlowFarneback(prvs, next, 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('Origin VS frame2', np.hstack([frame2, rgb]))
 cv2.imwrite('dof-imgs/' + str(num) + '.jpg', imutils.resize(np.hstack([frame2, rgb]), 600))
 k = cv2.waitKey(30) & 0xff
 num = num + 1
 if k == 27:
  break
 elif k == ord('s'):
  cv2.imwrite('dof-imgs/origin VS dense optical flow HSVres' + str(num) + ".jpg",
  imutils.resize(np.hstack([frame2, rgb]), width=800))
 prvs = next
cap.release()
cv2.destroyAllWindows()