Dimiter Kendri
Published © GPL3+

Voice Controlled AI Camera

Build a voice controlled AI camera that notifies with what it sees via email.

AdvancedFull instructions provided5 days3,314
Voice Controlled AI Camera

Things used in this project

Hardware components

Raspberry Pi 3 Model B
Raspberry Pi 3 Model B
×1
NVIDIA Jetson Nano Developer Kit
NVIDIA Jetson Nano Developer Kit
×1
Camera Module
Raspberry Pi Camera Module
×1

Software apps and online services

TensorFlow
TensorFlow
OpenCV
OpenCV

Story

Read more

Schematics

AI camera system conceptual block diagram

AI camera system conceptual block diagram

Code

TestSendEmailImage.py

Python
Tests sending images via email.
# Copyright dhq November 13 2019

# import all necessary components
import smtplib
from email.mime.text import MIMEText
from email.mime.image import MIMEImage
from email.mime.multipart import MIMEMultipart
import pickle
import socket
import struct
from PIL import Image
import numpy as np
import time
import os, sys
import sys
import cv2
import smtplib
import numpy as np

from imutils.video import VideoStream
import datetime
import argparse
import imutils
import time
import cv2
from time import sleep


HOSTNAME = 'smtp-mail.outlook.com'
USERNAME = ''
PASSWORD = ''
RECEPIENT_ADDRESS = ''
SUBJECT = 'Subject: AI camera using (imageNetV3) engine detected: '

""" Get email credentials
	:param none: 
	:returns none : 
"""
def LoginCredentials():
	USERNAME = input('Enter your email address: ')
	PASSWORD = input('Enter your email password: ')
	RECEPIENT_ADDRESS = input('Enter the recepient address: ')
	return USERNAME,PASSWORD,RECEPIENT_ADDRESS
	
	
def sendmail(detectedobject, user, emailpass, rcp_emailaddr):
	EMAIL_BODY = SUBJECT + str(detectedobject)
	
	msg = MIMEMultipart('related')
	msg['Subject'] = SUBJECT
	msg['From'] = USERNAME
	msg['To'] = RECEPIENT_ADDRESS

	html = """\
	<html>
	  <head></head>
		<body>
		  <img src="cid:image1" alt="Logo" style="width:300px;height:400px;"><br>
		   <p><h4 style="font-size:15px;">AI engine camera snapshot.</h4></p>           
		</body>
	</html>
	"""
	# Record the MIME types of text/html.
	part2 = MIMEText(html, 'html')

	# Attach parts into message container.
	msg.attach(part2)

	# This example assumes the image is in the current directory
	fp = open('ELE.JPG', 'rb')
	msgImage = MIMEImage(fp.read())
	fp.close()

	msgImage.add_header('Content-ID', '<image1>')
	msg.attach(msgImage)

	# print(msg.as_string())
	server = smtplib.SMTP(HOSTNAME, 587)
	server.ehlo()
	server.starttls()
	server.login(user, emailpass)	
	server.sendmail(user, [rcp_emailaddr],msg.as_string())
	server.quit()
	print ("Email sent. Logging Out...")
	
	
if __name__ == '__main__':
		try:
			print ('To exit the program press Ctrl-C.')
			USERNAME,PASSWORD,RECEPIENT_ADDRESS = LoginCredentials()
			detectedobject = str("found another lama")
			print ('Sending email \r\n.')
			sendmail(detectedobject, USERNAME, PASSWORD, RECEPIENT_ADDRESS)
			print ('Email sent \r\n.')

		except KeyboardInterrupt:
			print ('Quitting...')

ImgServerAI.py

Python
# Copyright dhq November 13 2019

# import all necessary components
import smtplib
from email.mime.text import MIMEText
from email.mime.image import MIMEImage
from email.mime.multipart import MIMEMultipart

import socket
import sys
import cv2
import pickle
import numpy as np
import struct ## new
import zlib

from PIL import Image
import os
import sys

# from skimage.io import imread
# from skimage.transform import resize
# import numpy as np
# from matplotlib import pyplot as plt
# from skimage import data, io
# from matplotlib import pyplot as plt
# from skimage import img_as_ubyte		#convert float to uint8
# from skimage.color import rgb2gray

from tensorflow.python.keras.preprocessing import image
from tensorflow.python.keras.applications.inception_v3 import *
import numpy as np

from imutils.video import VideoStream
import datetime
import argparse
import imutils
import time
import cv2
from time import sleep
from keras.models import load_model


HOST=''
PORT=8485


HOSTNAME = 'smtp-mail.outlook.com'
USERNAME = ''
PASSWORD = ''
RECEPIENT_ADDRESS = ''
SUBJECT = 'Subject: AI camera using (imageNetV3) engine detected: '

""" Get email credentials
	:param none: 
	:returns none : 
"""
def LoginCredentials():
	USERNAME = input('Enter your email address: ')
	PASSWORD = input('Enter your email password: ')
	RECEPIENT_ADDRESS = input('Enter the recepient address: ')
	return USERNAME,PASSWORD,RECEPIENT_ADDRESS
	
	
def sendmail(detectedobject, user, emailpass, rcp_emailaddr):
	EMAIL_BODY = SUBJECT + str(detectedobject)
	
	msg = MIMEMultipart('related')
	msg['Subject'] = SUBJECT
	msg['From'] = USERNAME
	msg['To'] = RECEPIENT_ADDRESS

	html = """\
	<html>
	  <head></head>
		<body>
		  <img src="cid:image1" alt="Logo" style="width:300px;height:400px;"><br>
		   <p><h4 style="font-size:15px;">AI engine camera snapshot.</h4></p>           
		</body>
	</html>
	"""
	# Record the MIME types of text/html.
	part2 = MIMEText(html, 'html')

	# Attach parts into message container.
	msg.attach(part2)

	# This example assumes the image is in the current directory
	fp = open('servpic.jpg', 'rb')
	msgImage = MIMEImage(fp.read())
	fp.close()

	msgImage.add_header('Content-ID', '<image1>')
	msg.attach(msgImage)

	# print(msg.as_string())
	server = smtplib.SMTP(HOSTNAME, 587)
	server.ehlo()
	server.starttls()
	server.login(user, emailpass)	
	server.sendmail(user, [rcp_emailaddr],msg.as_string())
	server.quit()
	print ("Email sent. ")
	
	
def main():
	model = InceptionV3(weights='imagenet')
	print('Deep CNN started')
	
	s=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
	print('Socket created')

	s.bind((HOST,PORT))
	print('Socket bind complete')
	s.listen(10)
	print('Socket now listening')

	conn,addr=s.accept()

	data = b""
	payload_size = struct.calcsize(">L")
	print("payload_size: {}".format(payload_size))
	while True:
		while len(data) < payload_size:
			print("Recv: {}".format(len(data)))
			data += conn.recv(4096)

		print("Done Recv: {}".format(len(data)))
		packed_msg_size = data[:payload_size]
		data = data[payload_size:]
		msg_size = struct.unpack(">L", packed_msg_size)[0]
		print("msg_size: {}".format(msg_size))
		while len(data) < msg_size:
			data += conn.recv(4096)
		frame_data = data[:msg_size]
		data = data[msg_size:]
		print(msg_size)
		frame=pickle.loads(frame_data)
		frame = cv2.imdecode(frame, cv2.IMREAD_COLOR)
		cv2.imwrite("servpic.jpg",frame)
		# cv2.imshow('ImageWindow',frame)
		# cv2.waitKey(1)
		img = image.load_img("servpic.jpg", target_size=(299, 299))
		x = image.img_to_array(img)
		x = np.expand_dims(x, axis=0)
		x = preprocess_input(x)
		y = model.predict(x)
		# cnt =  cnt + 1
		for index, res in enumerate(decode_predictions(y)[0]):
			print('{}. {}: {:.3f}%'.format(index + 1, res[1], 100 * res[2]))
		print(decode_predictions(y)[0])				#this is a list
		print(type(decode_predictions(y)[0][1]))	#this is a tuple
		print("Item detected : " + (decode_predictions(y)[0][0][1]))		# this picks the class description from the first tuple that shows the class with the highest probabililty
		decobj = (decode_predictions(y)[0][0][1])
		print("sending email to user\r\n")
		if(decobj == "")
			sendmail(decobj, USERNAME, PASSWORD, RECEPIENT_ADDRESS)


		
if __name__ == '__main__':
		try:
			print ('To exit the program press Ctrl-C.')
			USERNAME,PASSWORD,RECEPIENT_ADDRESS = LoginCredentials()
			print ('Email configuration done\r\n.')
			main()
		except KeyboardInterrupt:
			print ('Quitting...')

ImgClientTestCam.py

Python
# Copyright dhq November 13 2019

#sends camera frames to Jetson Nano

import cv2
import io
import socket
import struct
import time
import pickle
import zlib

JETSON_HOSTNAME = '192.168.2.13'		#erplace wih your own

client_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
client_socket.connect((JETSON_HOSTNAME, 8485))
connection = client_socket.makefile('wb')

cam = cv2.VideoCapture(0)

cam.set(3, 320);
cam.set(4, 240);

img_counter = 0

encode_param = [int(cv2.IMWRITE_JPEG_QUALITY), 90]

while True:
    ret, frame = cam.read()
    result, frame = cv2.imencode('.jpg', frame, encode_param)
#    data = zlib.compress(pickle.dumps(frame, 0))
    data = pickle.dumps(frame, 0)
    size = len(data)


    print("{}: {}".format(img_counter, size))
    client_socket.sendall(struct.pack(">L", size) + data)
    img_counter += 1

cam.release()

ImgClientMatrixAI.py

Python
#!/usr/bin/env python

# encoding: utf-8
from __future__ import unicode_literals
import cv2
import numpy as np
import socket
import sys
import pickle
import struct

# from skimage.io import imread
# from skimage.transform import resize
# import numpy as np
# from matplotlib import pyplot as plt
# from skimage import data, io
# from matplotlib import pyplot as plt
# from skimage import img_as_ubyte		#convert float to uint8
# from skimage.color import rgb2gray
from PIL import Image

# import the necessary packages
# from tensorflow.python.keras.preprocessing import image
# from tensorflow.python.keras.applications.inception_v3 import *
# from keras.models import load_model

from imutils.video import VideoStream
import datetime
import argparse
import imutils
import time
from time import sleep

import json
import subprocess

import paho.mqtt.client as mqtt

fromtimestamp = datetime.datetime.fromtimestamp

# MQTT client to connect to the bus
mqtt_client = mqtt.Client()
HOST = "localhost"
PORT = 1883
HOTWORD_DETECTED = "hermes/hotword/default/detected"
HOTWORDS_ON = {"lumos"}

		
# cap=cv2.VideoCapture(0)

# construct the argument parse and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument("-p", "--picamera", type=int, default=-1,
	help="whether or not the Raspberry Pi camera should be used")
args = vars(ap.parse_args())

# initialize the video stream and allow the cammera sensor to warmup
vs = VideoStream(usePiCamera=args["picamera"] > 0).start()
time.sleep(2.0)


# clientsocket=socket.socket(socket.AF_INET,socket.SOCK_STREAM)
# clientsocket.connect(('192.168.2.9',8089))

encode_param = [int(cv2.IMWRITE_JPEG_QUALITY), 90]


def main():
	img_counter = 0
	client_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
	client_socket.connect(('192.168.2.13', 8485))
	connection = client_socket.makefile('wb')
	print('Deep CNN started')
	
	# loop over the frames from the video stream
	while True:
		# grab the frame from the threaded video stream and resize it
		# to have a maximum width of 400 pixels
		frame = vs.read()
		frame = imutils.resize(frame, width=320)
	 
		# draw the timestamp on the frame
		timestamp = datetime.datetime.now()
		ts = timestamp.strftime("%A %d %B %Y %I:%M:%S%p")
		cv2.putText(frame, ts, (10, frame.shape[0] - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.35, (0, 0, 255), 1)
		
		# show the frame
		cv2.imshow("Frame", frame)
		
		result, frame = cv2.imencode('.jpg', frame, encode_param)
		#    data = zlib.compress(pickle.dumps(frame, 0))
		data = pickle.dumps(frame, 0)
		size = len(data)

		print("{}: {}".format(img_counter, size))
		client_socket.sendall(struct.pack(">L", size) + data)
		img_counter += 1


# Subscribe to the important messages
def on_connect(client, userdata, flags, rc):
    mqtt_client.subscribe(HOTWORD_DETECTED)


# Process a message as it arrives
def on_message(client, userdata, msg):
	if not msg.topic == HOTWORD_DETECTED:
		return

	payload = json.loads(msg.payload)
	model_id = payload["modelId"]
	# if model_id in HOTWORDS_ON:
		# #main()
		# else:
			# print("Unmapped hotword model_id: %s" % model_id)



if __name__=="__main__":
	main()
	mqtt_client.on_connect = on_connect
	mqtt_client.on_message = on_message
	mqtt_client.connect(HOST, PORT)
	mqtt_client.loop_forever()

Credits

Dimiter Kendri

Dimiter Kendri

23 projects • 151 followers
Robotics and AI

Comments

Related channels and tags
  • Displays
  • Machine Learning & AI
  • NVIDIA
  • OpenCV
  • Photos & Video
  • Raspberry Pi
  • Security
  • TensorFlow
  • Voice