OpenCV3计算机视觉Python语言实现（二）:处理文件、摄像头和图形用户界面
- 2.1 基本I/O脚本
- 2.1.1 读/写图像文件
- 2.1.2 图像和原始字节之间的转换
- 2.1.3使用numpy.array()访问图像数据
- 2.1.4 视频文件的读写
- 2.1.5 捕获摄像头的帧
- 2.1.6 在窗口显示图像
- 2.1.7 在窗口显示摄像头帧
- 2.2 Cameo项目（人脸跟踪和图像处理）
- 2.3 Cameo-面向对象设计
安装
从网址：http://www.lfd.uci.edu/~gohlke/pythonlibs/#opencv
下载opencv-whl文件
pip install wheel
pip install ****.whl大多数常用的OpenCV函数都在cv2模块中。可能会遇到其他基于cv2.cv模块的，这些都是传统版本，Python模块被成为cv2并不是表示该模块针对OpenCV2.x.x版本的，而是因为该模块引入了一个更好的API接口，它们采用了面向对象的编程方式，这与以前的cv模块有所不同，以前的cv模块更多采用面向过程化的编程方式。
2.1 基本的I/O脚本
2.1.1 读/写图像文件
OpenCV的imread()函数和imwrite()函数能支持各种静态图像文件。不管哪种格式，每个像素都会有一个值，但不同格式的表示像素的方式不同。如：通过numpy数组来创建一个黑色的正方形图像：
img=numpy.zeros((4,2),dtype=numpy.uint8)
# dtype=uint8表示每个像素都由一个8位整数来表示，即每个像素值的范围是0~255cv2.cvtColor函数：将该图像转换成BGR模式或者RGB模式
import cv2
imgBGR=cv2.cvtColor(img,cv2.COLOR_GRAY2BGR)
print imgBGR,imgBGR.shape
# img.shape():返回HxWxC.img为4行2列的二维数组。imgBGR.shape=(4,2,3),4页*2行*3列，其中每一个行的3列分别代表BGR值或BGR值.
取值img[0,0,0]或img[0][0]：第一个值表示y坐标或行，0表示顶部；第二个值表示像素的x坐标或者列；第三个值表示颜色通道
# 图像的读取、显示、保存
import cv2
img=cv2.imread('lena.jpg')
# 读取第二个参数表示以什么方式读入，将图像以灰度图像读入：cv2.imread('./lena.jpg',cv2.IMREAD_GRAYSCALE)第二个参数省略时，默认Opencv以BGR的格式读入
cv2.imshow('lena_img',img)
cv2.waitKey(0)
# cv2.waitKey(0)是一个键盘绑定函数。需要指出的是它的时间尺度是毫秒级。函数等待特定的几毫秒，看是否有键盘输入。特定的几毫秒之内，如果按下任意键，这个函数会返回按键的ASCII码，程序会继续运行。如果没有键盘输入，返回为-1，如果我们设置这个函数的参数为0，那么会无限期的等待键盘输入。它也可以被用来检测特定键是都被按下。
cv2.destoryAllWindows()
# 关闭所有打开的窗口。也可以通过cv2.destoryWindows(windowsName)来关闭指定的窗口
cv2.imwrite('lena2.png',img)
# imwrite()要求图像为BGR或者灰度格式2.1.2 图像和原始字节之间的转换:bytearray()函数
若一幅图像的通道有8位，则可将其显示转换成标准的一维Python bytearray格式：
import numpy as np
byteArray=bytearray(image)
print len(byteArray) # 输出为HxWxC
# 由字节转换成图像
grayImage=np.array(byteArray).reshape(height,width)
bgrImage=np.array(byteArray).reshape(height,width,3)
# 随机生成字节
import os
randomByteArray=bytearray(os.urandom(120000))
flatNumpyArray=numpy.array(randomByteArray) # OpenCV图像是.array类型的二维或者三维数组，因此转换成array"
bgrImage=flatNumpyArray.reshape(200,200,3)
cv2.imshow("bgrImage",bgrImage)2.1.3使用numpy.array()访问图像数据
访问数组中特定位置的值使用np.array()提供的item(),设置特定位置的值：itemset()
import cv2
import numpy as np
lenaImage = cv2.imread('lena.jpg')
print lenaImage.item(150, 120, 0)// 访问特定位置
lenaImage.itemset((150, 120, 0), 255) // 设置特定位置的值通过循环处理数组的效率很低，采用索引的方式来解决这个问题：
import cv2
import numpy as np
img = cv2.imread('lena.jpg') img[:,:,1]=0
img[:,:,1] = 0 // 将所有的G通道设置为0
img[0:100,0:100,:] // 得到图像从左上角开始100*100的区域三个通道值
img[0,0]=[0,0,0] // 将黑色BGR值赋给图像第0行第0列的三通道值其他属性
print img.shape // 返回图像H，W，C的数组
print img.size // 返回图像像素的大小
print img.dtype // 得到图像的数据类型（通常为一个无符号整数类型的变量和该类型占的位数，比如uint8）2.1.4 视频文件的读写
OpenCV提供了VideoCapture和VideoWriter类来支持各种视频的读写。在达到视频文件末尾之前，VideoCapture都会通过read()来获取新的一帧，每一帧都是基于BGR格式的图像；
可将一幅图像传递给VideoWriter类的write()函数，该函数会将这副图像加到VideoWriter类所指向的文件。
import cv2
videoCapture = cv2.VideoCapture('meipai.mp4')
fps = videoCapture.get(cv2.CAP_PROP_FPS) # 帧速率，fps帧/秒
size = (int(videoCapture.get(cv2.CAP_PROP_FRAME_WIDTH)), \
int(videoCapture.get(cv2.CAP_PROP_FRAME_HEIGHT)))
print size # (854,480)注意此时的size为WxH
videoWriter = cv2.VideoWriter('meipai.avi', cv2.VideoWriter_fourcc('I', '4', '2', '0'), fps, size)
success, frame = videoCapture.read()
print frame.shape # (480,854,3) 此时为读取到的每一张图片大小HxWxC
while success: # success为bool型，当读到末尾的时候为空
videoWriter.write(frame)
success, frame = videoCapture.read()
videoWriter类的构造函数需要指定视频编解码器。编解码器根据系统的不同而不同，一下是常用的一些选项：
- cv2.VideoWriter_fourcc('I','4','2','0'):该选项是一个未压缩的YUV颜色编码，是4:2:0色度子采样。这种编码有很好的兼容性，但会产生较大的文件，扩展名为avi
- cv2.VideoWriter_fourcc('P','I','M','1'):该选项是MPEG-1编码类型，扩展名为avi
- cv2.VideoWriter_fourcc('X', 'V', 'I', 'D'):MPEG-4编码，如果希望得到的视频大小为平均值，推荐使用此选项，扩展名avi
- cv2.VideoWriter_fourcc('T', 'H', 'E', 'O'):该选项是Ogg Vorbis,扩展名是avi
- cv2.VideoWriter_fourcc('F', 'L', 'V', '1'):该选项是一个Flash视频，扩展名为.flv2.1.5 捕获摄像头的帧
import cv2
cameraCapture = cv2.VideoCapture(0)
size = (int(cameraCapture.get(cv2.CAP_PROP_FRAME_WIDTH)), \
int(cameraCapture.get(cv2.CAP_PROP_FRAME_HEIGHT)))
fps = 30 # 假定帧速率为30
videoWriter = cv2.VideoWriter("MyOutputVideo.avi", \
cv2.VideoWriter_fourcc("I", "4", "2", "0"), fps, size)
success, frame = cameraCapture.read()
numFrameRemaining = 10 * fps - 1
while success and numFrameRemaining > 0:
videoWriter.write(frame)
success, frame = cameraCapture.read()
numFrameRemaining -= 1
cameraCapture.release()
2.1.6 在窗口显示图像
2.1.7 在窗口显示摄像头帧
import cv2
clicked = False
def onMouse(event, x, y, flags, param):
global clicked
if event == cv2.EVENT_LBUTTONUP:
clicked = True
cameraCapture = cv2.VideoCapture(0)
cv2.namedWindow("MyWindow") #指定窗口名
cv2.setMouseCallback("MyWindow", onMouse)#获取鼠标输入
print "Showing camera feed. Click window or press any key to stop."
success, frame = cameraCapture.read()
while success and not clicked and cv2.waitKey(1) == -1:#没达到停止条件时
cv2.imshow("MyWindow", frame)
success, frame = cameraCapture.read()
cv2.destroyAllWindows()
cameraCapture.release()
- cv2.waitKey():获取键盘输入，参数为等待键触发的时间，单位毫秒，其返回值为-1（表示没有被键盘按下）或ASCII码。
- setMouseCallback()获取鼠标输入
opencv的窗口函数和waitKey函数相互以阿里。Opencv的窗口只有在调用waitKey函数时才会更新，waitkey函数只有在opencv窗口成为活动窗口时才能捕获输入信息。
在一些系统中，waitKey()的返回值可能比ASCII码的值更大（在Linux系统中，如果Opencv使用GTK作为后端的GUI库，就会出现一个众所周知的bug
在所有的系统中，可以通过读取返回值的最后一个字节来保证只提取ASCII码，具体代码如下：
keycode = cv2.waitKey(1)
if keycode != -1:
keycode &= 0xFF2.2-2.3 cameo项目（人脸跟踪和图像处理）
1.manager.py
manager.py
# -*-coding:utf-8-*-
import cv2
from cv2 import *
import numpy
import time
class CaptureManager(object):
def __init__(self, capture, previewWindowManager=None, \
shouldMirrorPreview=False):
self.previewWindowManager = previewWindowManager
self.shouldMirrorPreview = shouldMirrorPreview
# 注意大多数成员（member）变量为非公有变量，这类变量名前会加一个下划线进行标识。
self._capture = capture
self._channel = 0
self._enteredFrame = False
self._frame = None
self._imageFilename = None
self._videoFilename = None
self._videoEncoding = None
self._videoWriter = None
self._startTime = None
self._framesElapsed = long(0)
self._fpsEstimate = None
# 关于@符号的含义可以参考链接：https://www.zhihu.com/question/26930016
@property
def channel(self):
return self._channel
@channel.setter
def channel(self, value):
if self._channel != value:
self._channel = value
self._frame = None
@property
def frame(self):
if self._enteredFrame and self._frame is None:
_, self._frame = self._capture.retrieve()
return self._frame
@property
def isWritingImage(self):
return self._imageFilename is not None
@property
def isWritingVideo(self):
return self._videoFilename is not None
def enterFrame(self):
"""Capture the next frame,if any."""
# But first ,check that any previous frame was exited.
assert not self._enteredFrame, 'previous enteredFrame() had no matching exitFrame()'
if self._capture is not None:
self._enteredFrame = self._capture.retrieve()
def exitFrame(self):
"""Draw to the window,write to files,Release the frame."""
# Check whether ant grabbed frame is retrievable.
# The getter may retrieve and cache the frame
if self._frame is None:
self._enteredFrame = False
return
# Update the FPS estimate and related variables
if self._framesElapsed == 0:
self._startTime = time.time()
else:
timeElapsed = time.time() - self._startTime
self._fpsEstimate = int(self._framesElapsed / timeElapsed)
self._framesElapsed += 1
# Draw to the Window,if any.
if self.previewWindowManager is not None:
if self.shouldMirrorPreview:
mirroredFrame = numpy.fliplr(self._frame).copy()
self.previewWindowManager.show(mirroredFrame)
else:
self.previewWindowManager.show(self._frame)
# Write to the image file,if any.
if self.isWritingImage:
cv2.imwrite(self._imageFilename, self._frame)
self._imageFilename = None
# Write to the video file,if any.
self._writeVideoFrame()
# Release the Frame
self._frame = None
self._enteredFrame = False
def writeImage(self, filename):
"""write the next exited frame to an image frame"""
self._imageFilename = filename
def startWritingVideo(self, filename, encoding=cv2.VideoWriter_fourcc("I", "4", "2", "0")):
"""start writing exited frames to a video file"""
self._videoFilename = filename
self._videoEncoding = encoding
print self._videoEncoding
def stopWritingVideo(self):
"""Stop writing exited frames to a video file"""
self._imageFilename = None
self._videoEncoding = None
self._videoWriter = None
def _writeVideoFrame(self):
if not self.isWritingVideo:
return
if self._videoWriter is None:
fps = self._capture.get(cv2.CAP_PROP_FPS)
if fps == 0.0:
# the capture fps is unknow ,so ues an estimate.
if self._framesElapsed < 20:
# wait until more frames elapse so that the estimate is stable
return
else:
fps = self._fpsEstimate
size = (int(self._capture.get(cv2.CAP_PROP_FRAME_WIDTH)), \
int(self._capture.get(cv2.CAP_PROP_FRAME_HEIGHT)))
# self._videoWriter = cv2.VideoWriter(self._videoFilename, self._videoEncoding, fps, size)
print "self._videoEncoding:",self._videoEncoding
self._videoWriter = cv2.VideoWriter(self._videoFilename, self._videoEncoding, fps, size)
print self._frame
self._videoWriter.write(self._frame)
class WindowManager(object):
def __init__(self, windowName, keypressCallback=None):
self.keypressCallback = keypressCallback
self._windowName = windowName
self._isWindowCreated = False
@property
def isWindowCreated(self):
return self._isWindowCreated
def createdWindow(self):
cv2.namedWindow(self._windowName)
self._isWindowCreated = True
def show(self, frame):
cv2.imshow(self._windowName, frame)
# cv2.waitKey(1)
def destroyWindow(self):
cv2.destroyWindow(self._windowName)
self._isWindowCreated = False
def processEvents(self):
keycode = cv2.waitKey(1)
if self.keypressCallback is not None and keycode != -1:
# Discard any non-ASCII info encoded by GTK
keycode &= 0xFF
self.keypressCallback(keycode)
2.cameo.py
import cv2
from managers import WindowManager, CaptureManager
import filters
class Cameo(object):
def __init__(self):
self._windowManger = WindowManager('Cameo', self.onKeypress)
self._captureManger = CaptureManager(cv2.VideoCapture(0), self._windowManger, True)
self._curveFilter=filters.BlurFilter()
def run(self):
"""Run the main loop"""
self._windowManger.createdWindow()
while self._windowManger.isWindowCreated:
self._captureManger.enterFrame()
frame = self._captureManger.frame
filters.strokeEdges(frame,frame)
self._curveFilter.apply(frame,frame)
self._captureManger.exitFrame()
self._windowManger.processEvents()
def onKeypress(self, keycode):
"""Handle a keypress.
space -> Take a screenshot
tab -> Start/stop recoding a screenshot
escape -> Quit
"""
if keycode == 32: # space
self._captureManger.writeImage("./screenshot.png")
elif keycode == 9: # tab
if not self._captureManger.isWritingVideo:
self._captureManger.startWritingVideo('./screenshot.avi')
else:
self._captureManger.stopWritingVideo()
elif keycode == 27: # escape
self._windowManger.destroyWindow()
if __name__ == '__main__':
Cameo().run()
这里出现一个问题，录制的视频大小不为0，但是时长总是0.不知道是什么原因？每一帧都被后一帧覆盖掉了？还不懂，还望懂的小伙伴不吝赐教。