OpenCV3 Python应用-2

1.视频文件读写

 1 # import cv2
 2 # videoCapture = cv2.VideoCapture('F:\\xzb\\Working\\5555.avi')
 3 # fps = videoCapture.get(cv2.CAP_PROP_FPS)
 4 # size = (int(videoCapture.get(cv2.CAP_PROP_FRAME_WIDTH)),
 5 #         int(videoCapture.get(cv2.CAP_PROP_FRAME_HEIGHT)))
 6 # videoWriter = cv2.VideoWriter(
 7 #     'F:\\xzb\\Working\\6666.avi',cv2.VideoWriter_fourcc("I","4","2","0"),
 8 #     fps,size)
 9 #
10 # success,frame=videoCapture.read()
11 # while success:
12 #     videoWriter.write(frame)
13 #     success,frame = videoCapture.read()

2.捕获摄像头的帧

 1 import cv2
 2 cameraCapture = cv2.VideoCapture(0)
 3 fps = 30
 4 #帧速率假设值
 5 size = (int(cameraCapture.get(cv2.CAP_PROP_FRAME_WIDTH)),
 6         int(cameraCapture.get(cv2.CAP_PROP_FRAME_HEIGHT)))
 7 videoWriter = cv2.VideoWriter(
 8     'F:\\xzb\\Working\\Made_video1.avi',cv2.VideoWriter_fourcc("I","4","2","0"),
 9     fps,size)
10 
11 success,frame=cameraCapture.read()
12 numFramesRemaining=10*fps-1
13 #设置时间
14 while success and numFramesRemaining >0:
15     videoWriter.write(frame)
16     success,frame = cameraCapture.read()
17     numFramesRemaining -=1
18 cameraCapture.release()    

cameraCapture类的get()方法不能返回摄像头帧速率的准确值,它总是返回0

常见于驱动程序仅支持基本功能的系统

要想针对摄像头构建合适的VideoWriter类,要么对帧速率作出假设,要么使用计时器来测量(推荐后者)

摄像头的数量和顺序由系统决定,但OpenCV不提供查询摄像头的数量和顺序的方法

如果使用无效索引构建了VideoCapture类,就不会得到帧,VideoCapture的read()返回(false,None)

为了避免这种情况,在执行该函数之后使用VideoCapture.isOpened方法做一个判断,返回一个Boolean值

当需要同步一组或者多头摄像头时,read()方法不再合适,可使用grab()和retrive()方法

　　如:

1 success0=cameraCapture0.grab()
2     success1=cameraCapture1.grab()
3     if success0 and success1:
4         frame0=cameraCapture0.retrive()

3.在窗口显示图像

1 import cv2
2 import numpy as np
3 
4 img = cv2.imread('test.jpg')
5 cv2.imshow('new_test.png',img)      #imshow()两个参数:显示图像的帧名称,要显示的图像本身
6 cv2.waitKey()
7 cv2.destroyAllWindows()              #释放由OpenCV创建的所有窗口

4.在窗口显示摄像头帧

namedWindow(),imshow(),DestoryWindow()函数允许指定窗口名来创建,显示,销毁窗口
任意窗口下都可以通过waitKey()函数获取键盘输入,通过setMouseCallback()函数获取鼠标输入
waitKey()的参数为等待键盘触发时间,单位为毫秒,其返回值是-1或者ASCII码(http://asciitable.com/)
    "注:有些系统,waitKey()返回值比ASCII偏大(LINUX中使用GTK为GUI库时有bug),可通过读取返回值最后一个字节提取ASCII码"
    如:

keycode = cv2.waitKey(1)
if keycode != -1:
    keycode &=0xFF

OpenCV的窗口函数和waitKey()函数相互依赖
窗口函数只有调用waitKey()才更新,waitKey()只有在OpenCV成为活动窗口才捕获输入信息
鼠标回调函数setMouseCallback()有五个参数,param为第三个是可选参数,默认是0
回调事件参数取不同的值对应不同的鼠标事件(具体见OpenCV3计算机视觉P.26)
OpenCV不提供任何处理窗口事件方法(如:点击窗口关闭按钮,无效),所以一般开发时将其嵌入应用程序框架中

 1 import cv2
 2 clicked = False
 3 def onMouse(event, x, y, flags, param):
 4     global clicked
 5     if event == cv2.EVENT_LBUTTONUP:
 6         clicked = True
 7 
 8 cameraCapture = cv2.VideoCapture(0)
 9 cv2.namedWindow('MyWindow')
10 cv2.setMouseCallback('MyWindow',onMouse)
11 
12 print('cbsujkbc jskb ')
13 success, frame = cameraCapture.read()
14 while success and cv2.waitKey(1) == -1 and not clicked:
15     cv2.imshow('MyWindow',frame)
16     success, frame = cameraCapture.read()
17 
18 cv2.destroyWindows('MyWindow')
19 cameraCapture.release()

---

---

格式啥的太乱了,还得熟悉在这写博客才行.