IP camera simulation using python and opencv

Question

I have a use-case in which I am simulating IP camera using python and opencv.

I am playing video using opencv and sending the bytes of the frame to an application which is streaming it at a port 8080.

The problem is as soon as the video gets finished, I have nothing to send to the application which is streaming this fake simulated camera at port 8080, so the application takes it as timeout and stops working.

My question is, how can I send some fake bytes lets days a black screen noise just to keep alive my application which is listening to my face simulated camera at 8080?

Edit 1: Adding code

app.py

from camera import VideoCamera
from flask import Flask, render_template, Response
import time

app = Flask(__name__)

@app.route('/')
def index():
    return render_template('index.html')

def gen(camera):
    while True:
        try:
            frame = camera.get_frame()
        except Exception:
            print("Video is finished or empty")
            #return None
            frame = camera.get_heartbeat()
        yield (b'--frame\r\n'
               b'Content-Type: image/jpeg\r\n\r\n' + frame + b'\r\n\r\n')

@app.route('/video_feed')
def video_feed():
    return Response(gen(VideoCamera()),
                    mimetype='multipart/x-mixed-replace; boundary=frame')

if __name__ == '__main__':
    app.run(debug=True)

camera.py

import cv2

class VideoCamera(object):
    def __init__(self):
        # Using OpenCV to capture from device 0. If you have trouble capturing
        # from a webcam, comment the line below out and use a video file
        # instead.
        #self.video = cv2.VideoCapture(0)
        # If you decide to use video.mp4, you must have this file in the folder
        # as the main.py.
        # self.video = cv2.VideoCapture('suits_hd.mp4') 
        self.video = cv2.VideoCapture('nature.mp4')

    def __del__(self):
        self.video.release()

    def get_frame(self):
        success, image = self.video.read()
        # We are using Motion JPEG, but OpenCV defaults to capture raw images,
        # so we must encode it into JPEG in order to correctly display the
        # video stream.
        ret, jpeg = cv2.imencode('.jpg', image)
        return jpeg.tobytes()

    def get_heartbeat(self):
        # jpeg = cv2.imread('noise-black.jpg')
        image = cv2.imread('noise-green.jpg')
        ret, jpeg = cv2.imencode('.jpg', image)
        return jpeg.tobytes()

If you show your code that sends the "normal" bytes, it should become clear how to send fake bytes. — Mark Setchell
– Mark Setchell, Commented Jul 21, 2018 at 10:19
So, when you init the VideoCamera, get the width and height of the video frames in the file and remember them. Then, if self.video.read() fails, just use numpy to create a random array the same size as a video frame and imencode() and send that. — Mark Setchell
– Mark Setchell, Commented Jul 21, 2018 at 11:14
` just use numpy to create a random array the same size as a video frame and imencode() and send that`, how to do this? — Prakash P
– Prakash P, Commented Jul 21, 2018 at 11:16
image = np.random.randint(0,256,(320,240,3), dtype=np.uint8) for a 320x240 frame in colour — Mark Setchell
– Mark Setchell, Commented Jul 21, 2018 at 11:18

Mark Setchell · Accepted Answer · 2018-07-21 11:46:55Z

1

So, when you init the VideoCamera, get the width and height of the video frames in the file and remember them. Then, if self.video.read() fails, just use numpy to create a random array the same size as a video frame and imencode() and send that.

Make a random green-ish frame with:

import numpy as np

# Make the Green channel out of intensities in range 200-255
G=np.random.randint(200,256,(320,240,1), dtype=np.uint8)
# Make Red and Blue channel out of intensities in range 0-49
X=np.random.randint(0,50,(320,240,1), dtype=np.uint8)
# Merge into a 3-channel image, use BGR order with OpenCV - although it won't matter because G is in the middle in both!
image=np.concatenate((X,G,X),axis=2)

edited Jul 21, 2018 at 11:46

answered Jul 21, 2018 at 11:41

Mark Setchell

210k32 gold badges310 silver badges505 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

avermaet · Accepted Answer · 2018-07-21 11:43:44Z

1

If you are able to recognize the end of the video file, you could just send a black image, which you could read from file. cv2.imread('black.jpg')and then send this one through the socket.

answered Jul 21, 2018 at 11:43

avermaet

1,59317 silver badges33 bronze badges

2 Comments

avermaet Over a year ago

For color green in the np array you'd need to generate the values with a coresponding rgb code.

Prakash P Over a year ago

I did this but the application reads a real ip camera (in my case reading a simulated fake ip camera) exit when feeded black image for more than 4 second. Green image works though.

Collectives™ on Stack Overflow

IP camera simulation using python and opencv

2 Answers 2

Comments

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related