pip install opencv-python
1. 无法修改摄像头参数
原因1：摄像头不支持此参数
比如：摄像头的帧率最高是30，但是你的电脑CPU带不动它，所以它在你电脑上的帧率只能达到22。
原因2：opencv的问题
我查到一个说法是opencv的权限比驱动低，所以无法修改，可以通过卸载摄像头驱动来修复，这个方法没有测试，看着就不靠谱。
因此就是opencv的问题了。
下面这个代码是广泛流传的打开摄像头代码，
import numpy as np
import cv2
cap = cv2.VideoCapture(0) # 打开摄像头
while(True):
ret, frame = cap.read() # 读取图片
cv2.imshow('frame',frame)
# 退出
if cv2.waitKey(1) & 0xFF == ord('q'):
break
cap.release()
cv2.destroyAllWindows()
代码是不存在问题，但是打开摄像头之后无法修改修改相机的参数，比如分辨率、帧率、亮度。
cap = cv2.VideoCapture(camera_number + cv2.CAP_DSHOW)
camera_number就是摄像头在你电脑上的设备号，0,1,2,…
CAP_DSHOW是opencv初始化的一个参数，我找到一个参数列表，适用参数应该与相机的驱动有关系，而我使用的笔记本的摄像头（USB相机），因此使用CAP_DSHOW就行了。
CAP_ANY = 0 # Auto detect
CAP_VFW = 200 # Video For Windows (obsolete, removed)
CAP_V4L = 200 # V4L/V4L2 capturing support
CAP_V4L2 = CAP_V4L # Same as CAP_V4L
CAP_FIREWIRE = 300 # IEEE 1394 drivers
CAP_FIREWARE = CAP_FIREWIRE # Same value as CAP_FIREWIRE
CAP_IEEE1394 = CAP_FIREWIRE # Same value as CAP_FIREWIRE
CAP_DC1394 = CAP_FIREWIRE # Same value as CAP_FIREWIRE
CAP_CMU1394 = CAP_FIREWIRE # Same value as CAP_FIREWIRE
CAP_QT = 500 # QuickTime (obsolete, removed)
CAP_UNICAP = 600 # Unicap drivers (obsolete, removed)
CAP_DSHOW = 700 # DirectShow (via videoInput)
CAP_PVAPI = 800 # PvAPI, Prosilica GigE SDK
CAP_OPENNI = 900 # OpenNI (for Kinect)
CAP_OPENNI_ASUS = 910 # OpenNI (for Asus Xtion)
CAP_ANDROID = 1000 # Android - not used
CAP_XIAPI = 1100 # XIMEA Camera API
CAP_AVFOUNDATION = 1200 # AVFoundation framework for iOS (OS X Lion will have the same API)
CAP_GIGANETIX = 1300 # Smartek Giganetix GigEVisionSDK
CAP_MSMF = 1400 # Microsoft Media Foundation (via videoInput)
CAP_WINRT = 1410 # Microsoft Windows Runtime using Media Foundation
CAP_INTELPERC = 1500 # RealSense (former Intel Perceptual Computing SDK)
CAP_REALSENSE = 1500 # Synonym for CAP_INTELPERC
CAP_OPENNI2 = 1600 # OpenNI2 (for Kinect)
CAP_OPENNI2_ASUS = 1610 # OpenNI2 (for Asus Xtion and Occipital Structure sensors)
CAP_GPHOTO2 = 1700 # gPhoto2 connection
CAP_GSTREAMER = 1800 # GStreamer
CAP_FFMPEG = 1900 # Open and record video file or stream using the FFMPEG library
CAP_IMAGES = 2000 # OpenCV Image Sequence (e.g. img_%02d.jpg)
CAP_ARAVIS = 2100 # Aravis SDK
CAP_OPENCV_MJPEG = 2200 # Built-in OpenCV MotionJPEG codec
CAP_INTEL_MFX = 2300 # Intel MediaSDK
CAP_XINE = 2400 # XINE engine (Linux)
这是我写的一个完整Demo
import cv2
import os
# Width, Height
Camera_PROP_640x480 = (640, 480)
Camera_PROP_800x600 = (800, 600)
Camera_PROP_960_640 = (960, 640)
Camera_PROP_1024x540 = (1024, 540)
Camera_PROP_1024x600 = (1024, 600)
Camera_PROP_1024x768 = (1024, 768)
Camera_PROP_1080x960 = (1080, 960)
Camera_PROP_1152x864 = (1152, 864)
Camera_PROP_1280x600 = (1280, 600)
Camera_PROP_1280x720 = (1280, 720)
Camera_PROP_1280x768 = (1280, 768)
Camera_PROP_1280x800 = (1280, 800)
Camera_PROP_1280x960 = (1280, 960)
Camera_PROP_1280x1024 = (1280, 1024)
Camera_PROP_1360x768 = (1360, 768)
Camera_PROP_1366x768 = (1366, 768)
Camera_PROP_1400x1050 = (1400, 1050)
Camera_PROP_1440x900 = (1440, 900)
Camera_PROP_1600x900 = (1600, 900)
Camera_PROP_1680x1050 = (1680, 1050)
Camera_PROP_1920x1080 = (1920, 1080)
Camera_PROP_2048x1080 = (2048, 1080)
Camera_PROP_4096x2160 = (4096, 2160)
Camera_PROP_Max = (300000, 300000)
# 解决显示图片过大导致图片跑到显示器外边
def imshow(winname, mat):
ratio = mat.shape[0] // 600
if ratio <= 1:
ratio = 1.5
cv2.namedWindow(winname, cv2.WINDOW_NORMAL)
cv2.resizeWindow(winname, int(mat.shape[1] / ratio), int(mat.shape[0] / ratio))
cv2.imshow(winname, mat)
if __name__ == '__main__':
print("=============================================")
print("= 热键(请在摄像头的窗口使用） =")
print("= w: 拍摄图片 =")
print("= q: 退出 =")
print("=============================================")
# 新建文件夹 test
class_name = 'test'
if not os.path.exists(class_name):
os.mkdir(class_name)
index = 1
# 打开相机
cap = cv2.VideoCapture(0 + cv2.CAP_DSHOW)
# fps = cap.get(cv2.CAP_PROP_FPS)
# 设置相机亮度
cap.set(cv2.CAP_PROP_BRIGHTNESS, 150)
print(cap.get(cv2.CAP_PROP_BRIGHTNESS))
# 设置相机增益
cap.set(cv2.CAP_PROP_GAIN, 1)
print(cap.get(cv2.CAP_PROP_GAIN))
# 设置相机分辨率
width, height = Camera_PROP_1280x720
cap.set(cv2.CAP_PROP_FRAME_WIDTH, width)
cap.set(cv2.CAP_PROP_FRAME_HEIGHT, height)
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
# 初始化opencv显示窗口‘capture’
cv2.namedWindow('capture', cv2.WINDOW_NORMAL)
cv2.resizeWindow('capture', width // 2, height // 2)
# 设置相机帧率
cap.set(cv2.CAP_PROP_FPS, 30)
print(cap.get(cv2.CAP_PROP_FPS))
# 新建一个'test.avi'文件
fps = 8
fourcc = cv2.VideoWriter_fourcc(*'XVID')
out = cv2.VideoWriter('test.avi', fourcc, fps, (width, height))
# 将相机录制到'test.avi'文件中
while (int(cv2.getWindowProperty('capture', 0)) != -1):
ret, frame = cap.read() # 获取当前帧
if not ret:
continue
frame = cv2.flip(frame, 1, dst=None) # 镜像
out.write(frame) # 写入帧
imshow("capture", frame) # 显示
# 判断相机退出
input = cv2.waitKey(1) & 0xFF
if input == ord('w'):
cv2.imwrite("%s/%d.png" % (class_name, index), frame)
print("%s: %d 张图片" % (class_name, index))
index += 1
elif input == ord('q'):
break
# 释放内存
out.release()
cap.release()
cv2.destroyAllWindows()
2. 中文无法识别
OpenCV的中文乱码问题暂时还无法解决，可以把系统编码修改为UTF8格式来解决这个问题，但是这个方法的代价可能太高了，因此只能使用一些“曲线救国”的办法。
2.1 窗口中文标题无法识别
方法一：改用C++的OpenCV
方法二：改用Python2版本，字符串变为 u’中文’
还是不建议使用中文标题
2.1 中文路径无法识别
import cv2
file_name = '测试/测试.bmp'
img = cv2.imread(file_name)
cv2.imshow('测试',img)
cv2.waitKey()
报错：
cv2.error: OpenCV(4.1.1) C:\projects\opencv-python\opencv\modules\highgui\src\window.cpp:352: error: (-215:Assertion failed) size.width>0 && size.height>0 in function 'cv::imshow'
方法一：
用np.fromfile读取，OpenCV再从内存中读取图片
import cv2
import numpy as np
file_name = '测试/测试.bmp'
img = cv2.imdecode(np.fromfile(file_name, dtype=np.uint8), -1)
cv2.imshow('test',img)
cv2.waitKey()
# 中文路径保存
cv2.imencode('.bmp', img)[1].tofile('测试/测试2.bmp')
方法二：
使用其他库读取图片，然后转成numpy格式再转成OpenCV格式，注意OpenCV是BGR
import cv2
import numpy as np
from PIL import Image
Image.MAX_IMAGE_PIXELS = 1000000000 # 图片最大内存
file_name = '测试/测试.bmp'
img_plt = Image.open(file_name)
# pillow转cv
img_cv = cv2.cvtColor(np.asarray(img_plt), cv2.COLOR_RGB2BGR)
# cv转pillow
img_plt = Image.fromarray(cv2.cvtColor(img_cv, cv2.COLOR_BGR2RGB))
cv2.imshow('test', img_cv)
cv2.waitKey()
方法三：
这个方法跟上一个方法类似，但是我觉得比较有意思，是使用urllib通过url来获取图片，接着转成numpy格式再转为OpenCV格式
import cv2
import numpy as np
from urllib.request import urlopen,pathname2url
file_name = '测试/测试.bmp'
file_url = 'file:' + pathname2url(file_name)
img_url = urlopen(file_url)
img_np = np.asarray(bytearray(img_url.read()), dtype="uint8")
img_cv = cv2.imdecode(img_np, cv2.IMREAD_COLOR)
cv2.imshow('test', img_cv)
cv2.waitKey()
3. 图片转BGR
opencv貌似除了cv2.imread(file_name, 0)外，没有自动匹配位深度转灰度，因此在将图片转灰度需要考虑位深度，否则会报错
def convert_BGR(image):
"""
转BGR
:param image:原图像
"""
if len(image.shape) == 2:
image = cv2.cvtColor(image, cv2.COLOR_GRAY2BGR)
elif len(image.shape) == 3:
if image.shape[2] == 3:
return image
elif image.shape[2] == 4:
image = cv2.cvtColor(image, cv2.COLOR_BGRA2BGR)
else:
assert 'image channel too many'
else:
assert 'image channel too many'
return image
4. resize问题
获取图片尺寸用img.shape()，会得到(高，宽，通道数)，单通道图片是(高，宽)
但是得到的resize图片的顺序是(宽，高)
img = cv2.resize(img, (width,height))
如果修改图片尺寸，用OpenCV的imshow显示正常，但是传到其他控件考虑会产生裂图（图片显示不正常）
这是因为OpenCV的显示使用了优化，而其他控件打开图片是遵守width，height满足4的倍数
比如PyQt的Qlabel控件显示图片需要将OpenCV的图片转换成QImage，不仅仅是传图片，还要对图片的尺寸进行调整，代码如下：
showImage = QtGui.QImage(show.data, show.shape[1], show.shape[0], show.shape[1] * show.shape[2], QtGui.QImage.Format_RGB888)
