recording.md

Capturing Video at Regular Intervals

The following example code can be run on the camera pi to capture and save video at regular intervals. The captured footage is stored locally on the Pi and must be manually transferred to the device running the AI back end. In the below example, the footage is stored on a USB drive mounted on `/mnt/usb/.

from picamera import PiCamera
from time import sleep
from datetime import datetime
 
picamera = PiCamera()
picamera.resolution = (720, 480)
picamera.framerate = 1
 
n = 0 # keeps track of the number of clips saved

RECORDING_INTERVAL = 15 # the number of minutes that the camera records for at each interval
 
while(True):
        # check to see if it is between the hours of 7:00 am and 5:30 pm
        time = str(datetime.now().time())[0:5].split(':')
        #print(time)
        if int(time[0]) > 7 and int(time[0]) < 18:
                # time to record!
                print('recording...')
                picamera.start_recording('/mnt/usb/clip'+str(n)+'_{0}.h264'.format(datetime.now().strftime("%Y-%m-%d")))
                sleep(RECORDING_INTERVAL*60)
                picamera.stop_preview()
                picamera.stop_recording()
                sleep(3600 - RECORDING_INTERVAL*60)
                n += 1
        else:
                # wait
                print('not a good time, waiting...')
                sleep(3600) # wait an hour

This script can be configured to run on boot using systemctl. Save the above script and take note of the file location. Create a systemd configuration file for this script using the following command:

sudo nano /etc/systemd/system/local_video_capture.service

Add the following content:

[Unit]  
Description=Captures video at regular intervals and saves the footage locally
After=multi-user.target

[Service]  
Type=simple  
Restart=always  
ExecStart=/usr/bin/python3 /location/of/the/above/script.py

[Install]  
WantedBy=multi-user.target

Save the file, and run the following command to make sure the script will run successfully on boot:

sudo systemctl daemon-reload
sudo systemctl enable local_video_capture.service
sudo systemctl start local_video_capture.service
sudo systemctl status local_video_capture.service

If the above command return that the script is running without error, it is now configured to run when the Pi boots.

Offloading the Video

After deployment, the captured video must be offloaded from the Pi in order to be run through the AI back end. This can be done using the scp command.

Begin by ssh-ing into the directory where the above script has been storing video. Use the zip command to compress the captured video:

zip recordings.zip *.h264

Next, use scp to transfer the zip file to the device ssh-ing into the Pi. Run the following command on this other device:

scp [username]@[hostname].local:where/the/recordings/are/stored/on/the/pi/recordings.zip /path/to/save/file

Running the Captured Footage Through the AI Back End

The back end can be configured as follows to process the collected footage (where the above zip file has been extracted to ./clips/Deployment):

try:
    # iterate over each frame in the video file captured locally on the camera pi
    for clip in sorted(os.listdir('./clips/Deployment')):
        video = './clips/Deployment/' + clip
        cap = cv2.VideoCapture(video)
        clip_date = clip.split('_')[1].split('.')[0] # just get the date from file name of the form clipN_[date].h264

        # iterate over the frames
        while cap.isOpened():
            # read the current frame
            ret, frame = cap.read()

            if not ret:
                break

            frame_image = Image.fromarray(frame)
            frame_image.convert('RGB').save('fish.png')
            # rotate the image
            #frame_image = frame_image.rotate(90)
            run_model(frame_image)

        # release the video capture object
        cap.release()
except KeyboardInterrupt:
    print('exiting...')