所以我要这样做:
fourcc = cv2.VideoWriter_fourcc(*'avi')
source = cv2.VideoCapture('video.avi')
while(source.isOpened()):
ret, frame = source.read()
if(type(frame) != type(None)):
for line in frame:
for pixel in line:
someedit()
我的问题是,5秒的300:400视频需要大约5分钟,即使someedit()是一些基本的像像素[0]+1。python做这样的事情通常很慢吗,还是有解决办法?
#import deps
import cv2
import numpy as np
#read the video
source = cv2.VideoCapture('videoPath')
# flag
ret = True
# iterate through all the frames if the video clip present
while(source.isOpened() and ret):
#read the frame
ret, frame = source.read()
#check for flag
if(ret):
# add 1 to 0th channel
frame[:, :, 0] = 1 + frame[:, :, 0]
print("Done")