I have obtained the frame_count of a saved video.
self.frame_count = self.capture.get(cv.CV_CAP_PROP_FRAME_COUNT) - 1
Now, I want to start a frame read from a particular frame_count. How do I do this?
Reason: I need to track an object and I have found the location of the object I want to track using HSV image segmentation. Now to track it, I intend to start the video from that particular frame and set the track window to the objects' coordinates.
Want: It should not be redundant and computationally intensive.
Use the below code to accomplish your task.
The Opencv version you are using is old. Use Opencv2.
Link : http://docs.opencv.org/trunk/doc/py_tutorials/py_tutorials.html
import numpy as np
import cv2
import os
cam = cv2.VideoCapture('full_path/bird.avi')
i = 1
initialize_first_frame = cam.read()[1]
start_from_frame = 5
dir = './testImages/'
os.makedirs(dir)
while True:
if(i>=start_from_frame):
cv2.imshow('Frame Conversion',cam.read()[1])
cv2.imwrite(dir + "/image-"+str(i).zfill(5)+".png",cam.read()[1])
i = i + 1
key = cv2.waitKey(10)
if key == 27:
cv2.destroyWindow('Frame Conversion')
break
print "End"
I hope this is the code you want.
Try the following:
f = # put here the frame from which you want to start
self.capture.set(CV_CAP_PROP_POS_FRAMES, f)
while f < self.frame_count:
retval, image = self.capture.read()
f += 1
For any version 3 and 4 of OpenCV Python, follow this -
import cv2
# suppose you want to start reading from frame no 500
frame_set_no = 500
cap = cv2.VideoCapture("full_path/cow.mp4")
# capture is set to the desired frame
cap.set(cv2.CAP_PROP_POS_FRAMES, frame_set_no)
while True:
ret, frame = cap.read()
cv2.imshow("Video", frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
cv2.destroyAllWindows()
Related
I wish to write a program for face recognition. I want my program to automatically take a specific number of images(say,5). Here is what I am currently doing:
import cv2
cam=cv2.VideoCapture(0)
a=0
count=0
for a in range(5):
ret,img=cam.read()
cv2.imshow("Test",img)
if not ret:
break
k=cv2.waitKey(1)
print("Image"+str(count)+"saved")
file='C:/Users/User/Desktop/FACEREC/known_faces/img_test/ad2'+str(count)+'.jpg'
cv2.imwrite(file,img)
a+=1
cam.release()
cv2.destroyAllWindows()
cam.release()
cv2.destroyAllWindows()
It shows "Image 0 saved" five times but in the folder(ad2), there is only a single image. Need some guidance.
Thank you
Lets start interpreting the code line-by-line:
Variable initialization
import cv2
cam=cv2.VideoCapture(0)
a=0
count=0
You load the opencv library and initialize the cam variable for webcam display. Then both a and count are intialized to 0. Now count variable is understandable, most probably count count the frames. What does a do?
Display and save frames
First-part
for a in range(0, 5):
ret,img=cam.read()
cv2.imshow("Test",img)
if not ret:
break
variable a declared again so we don't need to intialize at the beginning. Then current web-frame is being fetched. The frame is given as an input to imshow without even the current frame returned or not. The correct logic should be:
for a in range(5):
ret,img=cam.read()
if ret:
cv2.imshow("Test",img)
Second-part
k=cv2.waitKey(1)
print("Image"+str(count)+"saved")
file='C:/Users/User/Desktop/FACEREC/known_faces/img_test/ad2'+str(count)+'.jpg'
cv2.imwrite(file,img)
a+=1
Now, the k variable is initialized as a key for quiting the program when a key is pressed. Say, when q is pressed program terminates.
k=cv2.waitKey(1) & 0xFF
if k == ord('q'):
break
But if you break during the process, then you can't save the images. Therefore break should be at the end of the line.
For the rest of the code, you can use a as a counter. If you specify the start and end values. For each value a is increased by one. Then you can use it to discriminate the frames.
Code:
# Load the library
import cv2
# Initialize webcam
cam = cv2.VideoCapture(0)
# Save the frames
for a in range(0, 5):
# Get current frame
ret, img=cam.read()
# If frame is returned
if ret:
# Set the file-name
file='C:/Users/User/Desktop/FACEREC/known_faces/img_test/ad2' + str(a) + '.jpg'
# Write the frame
cv2.imwrite(file, img)
# Inform the user
print("Image" + str(a) + "saved")
# Display the frame
cv2.imshow("Test", img)
# If q is pressed quit
k=cv2.waitKey(1) & 0xFF
if k == ord('q'):
break
# Release the object
cam.release()
# Close the window
cv2.destroyAllWindows()
Here is the code for the same, have a look at it. In this, below code I am creating a Motion Detector and with this I will be recording the timings of when the various objects appeared and disappeared for which I am using dataframe.
The issue with this is that the program executes but when the output is displayed in the form of a Window, it freezes when I try to terminate.
import cv2, pandas
from datetime import datetime
first_frame = None
status_list = [None,None]
times = []
df = pandas.DataFrame(columns=["Start", "End"]) #Dataframe to store the time values during which object detection and movement appears.
video = cv2.VideoCapture(0)#VideoCapture object is used to record video using web cam
while True:
#check is a bool data type, returns true if VideoCapture object is read
check,frame = video.read()
status = 0
gray = cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY) # For converting the frame color to gray scale
gray = cv2.GaussianBlur(gray,(21,21),0) # For converting the gray scale frame to GaussianBlur
if first_frame is None:
first_frame = gray # used to store the first image/frame of the video
continue
delta_frame = cv2.absdiff(first_frame,gray)#calculates the difference between first and other frames
thresh_delta = cv2.threshold(delta_frame,30,255,cv2.THRESH_BINARY)[1]
thresh_delta = cv2.dilate(thresh_delta,None,iterations=0) #Provides threshold value, so if the difference is <30 it will turn to black otherwise if >30 pixels will turn to white
(_,cnts,_) = cv2.findContours(thresh_delta.copy(),cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE) #Define the contour area,i.e. adding borders
#Removing noises and shadows, any part which is greater than 1000 pixels will be converted to white
for contour in cnts:
if cv2.contourArea(contour) < 1000:
continue
status = 1 #change in status when the object is detected
#Creating a rectangular box around the object in frame
(x,y,w,h) = cv2.boundingRect(contour)
cv2.rectangle(frame,(x,y),(x+w,y+h),(0,255,0),3)
#list of status for every frame
status_list.append(status)
status_list=status_list[-2:]
#Record datetime in a list when change occurs
if status_list[-1]==1 and status_list[-2]==0:
times.append(datetime.now())
if status_list[-1]==0 and status_list[-2]==1:
times.append(datetime.now())
cv2.imshow('frame',frame)
#cv2.imshow('Capturing',gray)
#cv2.imshow('delta',delta_frame)
#cv2.imshow('thresh',thresh_delta)
key = cv2.waitKey(1) #changing the frame after 1 millisecond
#Used for terminating the loop once 'q' is pressed
if key == ord('q'):
break
print(status_list)
print(times)
for i in range(0,len(times),2):
df = df.append({"Start":times[i],"End":times[i+1]},ignore_index=True)
df.to_csv("Times.csv")
video.release()
cv2.destroyAllWindows #will be closing all the windows
Try this:
if cv2.waitKey(1) & 0xFF == ord('q'):
break
For a brief explanation about what "& 0xFF" is, see: What's 0xFF for in cv2.waitKey(1)?
It is basically a bit mask that takes the portion of 'waitKey' value(32 bit) that can be compared to ASCII (8 bit).
You can try this.
k = cv2.waitKey(1) & 0xFF
if k == 27 :
break
where k can be any ASCII code on the keyboard ASCII code help
In this case 27 is 'Esc' button.
However, the problem you're having where it is freezing, have you tried continuously pressing any button besides q? I had a similar problem when I was testing out OpenCV's tutorial code here
but i figured it out by playing around and then realizing i needed to hold down any button besides the 'exit' button.
I had faced the same issue bcz of this piece of code from geeksforgeeks
Please chck the last line of the code:
cv2.destroyAllWindows #will be closing all the windows
Add parenthesis, it shud b:
cv2.destroyAllWindows()
Please try this, rest looks fine
Well you can also pass a particular frame that you want to close like:
cv2.destroyAllWindows("frame")
I am running the following code:
import numpy as np
import cv2
import os
count = 0
cap = cv2.VideoCapture("/home/simon/PROJECT/real_data/00000020.mp4")
while not cap.isOpened():
cap = cv2.VideoCapture("./00000020.mp4")
cv2.waitKey(1000)
print "Wait for the header"
pos_frame = cap.get(cv2.cv.CV_CAP_PROP_POS_FRAMES)
while True:
flag, frame = cap.read()
if flag:
# The frame is ready and already captured
cv2.imshow('video', frame)
pos_frame = cap.get(cv2.cv.CV_CAP_PROP_POS_FRAMES)
print str(pos_frame)+" frames"
else:
# The next frame is not ready, so we try to read it again
cap.set(cv2.cv.CV_CAP_PROP_POS_FRAMES, pos_frame-1)
print "frame is not ready"
# It is better to wait for a while for the next frame to be ready
cv2.waitKey(1000)
if cv2.waitKey(10) & 0xFF == 27:
break
if cap.get(cv2.cv.CV_CAP_PROP_POS_FRAMES) == cap.get(cv2.cv.CV_CAP_PROP_FRAME_COUNT):
# If the number of captured frames is equal to the total number of frames,
# we stop
break
if ret == True:
frame = cv2.VideoCapture.grab()
frame = 'frame%d' % count
cv2.imwrite('frame%d.png', frame)
count += 1
else:
print 'stopped at' + count
break
And whenever I run it, it loops on the while not loop, printing "wait for header".
There is never an error code or anything like that either.
I have tried to run it as a more simple piece of code, where it doesnt have all these checks, and again that doesn't throw any errors.
I am attempting to run this code to open a video, and then save the frames as png files throughout the video.
Does anyone spot any particular problems with the code?
Or alternatively does anyone know a piece of code that would do what i want more efficiently, as I have trawled through google searches and stack overflow a lot recently and haven't found anything
Thanks in advance
Panda
You need to include a couple of DLLs in your python directory in order to play videos. Please see this for details:
https://li8bot.wordpress.com/2014/07/09/opencvpython-part-1-working-with-videos/
I need to compare 2 videos to check whether they are the same.
So I am planning to do the following:
Split both the videos into individual frames
Compare each frame with the corresponding frame of the ref video using Python Image Lib
Count the number of different frames to decide whether they are same.
I would like to know whether there is any function in Python to help me with the first step, i.e., to split the video into individual frames. I do not want to use ffmpeg to do the splitting.
Thanks in advance for the help
You can use opencv
import cv2
video_capture = cv2.VideoCapture("rtsp://admin:admin#192.168.0.94:554/stream3")
while True:
# get frame by frame
ret, frame = video_capture.read()
cv2.imwrite('pic.png',frame)
cv2.imshow('Video', frame)
Try this:
currentFrame = 0
while(True):
# Capture frame-by-frame
ret, frame = cap.read()
# Saves image of the current frame in jpg file
name = './data/frame' + str(currentFrame) + '.jpg'
print ('Creating...' + name)
cv2.imwrite(name, frame)
# To stop duplicate images
currentFrame += 1
# When everything done, release the capture
cap.release()
cv2.destroyAllWindows()
#you can use this method
I'm trying to have my webcam do the following:
capture 10 captures for 1 minute duration (one capture every 6
seconds).
Then save them into an array or on the computer.
Then get other captures for 5 minutes duration (one capture every 6
seconds) and compare each capture with the first array.
so far I have this code:
import cv
cv.NamedWindow("Camera", 1)
# in case im using another camera:
cv.NamedWindow("Camera", 2)
capture = cv.CaptureFromCAM(0)
img = cv.QueryFrame(capture)
while True:
img = cv.QueryFrame(capture)
cv.ShowImage("Camera", img)
if cv.WaitKey(10) == 27:
break
cv.DestroyWindow("Camera")
How do I proceed with this? any ideas?
thanks in advance.