main.py

'''
This is main function, used to start instances of the full system
'''
from arlo import Arlo
from subprocess import call
from shared_variables import Shared_Variables
from camera_handler import *
from vizualise import Vizualise
from vizualise_detections import Vizualise_Detections
from obj_detection import Obj_Detection
from threading import Thread
from pyfiglet import Figlet
import os, re
from monitor import *

"""
Change These Variables, some functions won't work while running in docker!
"""
source_list = [0]

# Set both these to False in order to use default image sizes
UTILIZE_ALL_MONITOR_SPACE = True # Set this to maximize visual size of all images
VISUALIZATION_OFFSET_AND_SIZE = False # If UTILIZE_ALL_MONITOR_SPACE is false, use this size instead, set to False if you want all screen to load on eachother with camera default size.
MONITOR_INDEX_LIST = False # fill this list with the index numbers of monitors to use [1 2 3]

SHOW_ALL_PERSONS = False # Decide if you want to show all cropped images of detected persons

use_arlo = False # Set to True if you want to scan for and use arlo cameras!
USERNAME = '' # for arlo login
PASSWORD = ''

DETECTION_OPTIMIZE_SIZE = [640,480] # Set size of images before prediction, Set to None if you want to use full size images

"""
----- PROGRAM CODE -----
"""
path = "/proc/" + str(os.getpid()) + "/cgroup"

def is_docker():
  if not os.path.isfile(path): return False
  with open(path) as f:
    for line in f:
      if re.match("\d+:[\w=]+:/docker(-[ce]e)?/\w+", line):
        return True
    return False


def create_instances( source_list , monitor_list):
    '''
    Create instances of system
    @param get list of sources, int for webcams and str for rtsp
    '''
    camera_amount = len(source_list)
    shared_variables = Shared_Variables(amount = camera_amount)
    if not is_docker():
        monitor_box_list = split_monitors(monitor_list, source_list, MONITOR_INDEX_LIST)

    # start show all detection crop images thread
    if SHOW_ALL_PERSONS:
        Vizualise_Detections(shared_variables=shared_variables).start()

    for i in range(0, camera_amount):

        # Create camera stream
        cam = Camera(id = i, cam_id=source_list[i], shared_variables=shared_variables, DETECTION_OPTIMIZE_SIZE=DETECTION_OPTIMIZE_SIZE).start()

        # Start Object detection on camera
        Obj_Detection( id = i, shared_variables=shared_variables, DETECTION_OPTIMIZE_SIZE=DETECTION_OPTIMIZE_SIZE).start()

        # Set Visualization mode
        if UTILIZE_ALL_MONITOR_SPACE:
            Vizualise(id = i, shared_variables=shared_variables, pos = monitor_box_list[i]).start()
        elif VISUALIZATION_OFFSET_AND_SIZE:
            Vizualise(id = i, shared_variables=shared_variables, pos = VISUALIZATION_OFFSET_AND_SIZE).start()
        else:
            Vizualise(id = i, shared_variables=shared_variables).start()

        print("Created instance ", i," with camera, detection and vizualisation.")

# Main start here
if __name__ == "__main__":
    f = Figlet(font='slant')
    print (f.renderText('Number of Persons in Room'))
    print("Please hold on while we receive cameras and start detection threads...")

    if is_docker():
        print("Running on docker!")
        UTILIZE_ALL_MONITOR_SPACE = False # Set this to maximize visual size of all images
        VISUALIZATION_OFFSET_AND_SIZE = False # If UTILIZE_ALL_MONITOR_SPACE is false, use this size instead, set to False if you want all screen to load on eachother with camera default size.
        MONITOR_INDEX_LIST = False # fill this list with the index numbers of monitors to use [1 2 3]
        monitor_list = None
    else:
        print("Collecting Monitor information...")

        print("----- Print Monitor List -----")
        monitor_list = detect_monitors()
        print(monitor_list)
        print("----- Monitor List End -----")

    print("Setting up source_list...")

    print("Adding local cameras...")

    if use_arlo:
        print("Waiting for Arlo cameras...")
        try:

            # Instantiating the Arlo object automatically calls Login(),
            # which returns an oAuth token that gets cached.
            # Subsequent successful calls to login will update the oAuth token.
            arlo = Arlo(USERNAME, PASSWORD)
            # At this point you're logged into Arlo.

            # Get the list of devices and filter on device type to only get the cameras.
            # This will return an array which includes all of the canera's associated metadata.
            cameras = arlo.GetDevices('camera')

            # Get the list of devices and filter on device type to only get the basestation.
            # This will return an array which includes all of the basestation's associated metadata.
            basestations = arlo.GetDevices('basestation')

            # Open the event stream to act as a keep-alive for our stream.
            arlo.Subscribe(basestations[0])

            # Send the command to start the stream and return the stream url.
            #url = arlo.StartStream(basestations[0], cameras[0])

            i = 0

            for cam in cameras:
                url = arlo.StartStream(basestations[0], cameras[i])
                source_list.append(url)

                i+=1

        except Exception as e:
            print(e)

    print("Done with collecting cameras...")
    print("----- Print source list -----")
    for address in source_list:
        print(address)
    print("----- Source list end -----")

    print("Creating all instances... this might take a while!")
    create_instances(source_list, monitor_list)