在检测出运动的物体之后，我还需要知道运动的方向，使用了上一节中的办法检测运动我发现很难去计算运动方向，开始考虑通过计算轮廓的中点的变化来实现，但是因为每次检测出得轮廓的数量不稳定，所以这个办法会让误差不可控。

这时我发现了goodFeaturesToTrack函数，简直是救了我，goodFeaturesToTrack函数可以获取图像中的最大特征值的角点，以下是我的思路：

• 对两帧图像做一个absdiff得到新图像。

• 对新图像做灰度和二值化处理。

• 使用goodFeaturesToTrack函数得到最大特征值的角点。

• 计算角点的平均点，扔进队列。

• 维护一个长度为10的队列，队列满时计算队列中数据的增减情况，来确定运动方向。

以下是代码示例：

#!usr/bin/env python
#coding=utf-8import cv2
import numpy as np
import Queuecamera = cv2.VideoCapture(0)
width = int(camera.get(3))
height = int(camera.get(4))firstFrame = None
lastDec = None
firstThresh = Nonefeature_params = dict( maxCorners = 100,qualityLevel = 0.3,minDistance = 7,blockSize = 7 )# Parameters for lucas kanade optical flow
lk_params = dict( winSize  = (15,15),maxLevel = 2,criteria = (cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10, 0.03))color = np.random.randint(0,255,(100,3))
num = 0q_x = Queue.Queue(maxsize = 10)
q_y = Queue.Queue(maxsize = 10)while True:(grabbed, frame) = camera.read()gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)gray = cv2.GaussianBlur(gray, (21, 21), 0)if firstFrame is None:firstFrame = graycontinueframeDelta = cv2.absdiff(firstFrame, gray)thresh = cv2.threshold(frameDelta, 25, 255, cv2.THRESH_BINARY)[1]# 下面的是几种不同的二值化的方法，感觉对我来说效果都差不多# thresh = cv2.adaptiveThreshold(frameDelta,255,cv2.ADAPTIVE_THRESH_GAUSSIAN_C,\# cv2.THRESH_BINARY,11,2)# thresh = cv2.adaptiveThreshold(frameDelta,255,cv2.ADAPTIVE_THRESH_MEAN_C,\#           cv2.THRESH_BINARY,11,2)thresh = cv2.dilate(thresh, None, iterations=2)p0 = cv2.goodFeaturesToTrack(thresh, mask = None, **feature_params)if p0 is not None:x_sum = 0y_sum = 0for i, old in enumerate(p0):x, y = old.ravel()x_sum += xy_sum += yx_avg = x_sum / len(p0)y_avg = y_sum / len(p0)if q_x.full():# print list(q_x.queue)qx_list = list(q_x.queue)key = 0diffx_sum = 0for item_x in qx_list:key +=1if key < 10:diff_x = item_x - qx_list[key]diffx_sum += diff_x# print diff_xif diffx_sum < 0 and x_avg < 500:# print "some coming form left"cv2.putText(frame, "some coming form left", (100,100), 0, 0.5, (0,0,255),2)else:print "right"print x_avgq_x.get()q_x.put(x_avg)cv2.putText(frame, str(x_avg), (300,100), 0, 0.5, (0,0,255),2)frame = cv2.circle(frame,(int(x_avg),int(y_avg)),5,color[i].tolist(),-1)cv2.imshow("Security Feed", frame)firstFrame = gray.copy()camera.release()
cv2.destroyAllWindows()

总的来讲作为一个图像处理的小白，不断地折腾和尝试，终于搞出了自己想要的东西，OpenCV绝对是喜欢折腾的人必要掌握的一个库了，以后肯定还会继续研究这块东西。