simplecv_final.py

'''THIS IS A MOTION DETECTING PROGRAM FOR THE RASPBERRY PI
It is part of a tutorial series that you can find here:
https://www.youtube.com/playlist?list=PLlg8lN4r9qWiDzF13lJY-lGtiTFHHGcpx
Running this program requires installing SimpleCV as well as
a few other prerequisits on your pi. You can find detailed
instructions for how to do that here:
http://tinkernut.com/YtQH9
'''

#!/usr/bin/python

#import the SimpleCV, shutil and the custom  py_gmailer  libraries
from SimpleCV import *
import py_gmailer
import shutil

#initialize the camer
cam = Camera()
#set the max display size
display = Display((800,600))

#create a threshold variable to change  motion sensitivity
threshold = 5.0

#set timer variables for email loop
start_time = time.time()
wait_time = 60 #in seconds

#set a streaming variable to stream webcam online
streaming = JpegStreamer("0.0.0.0:1212")

#create destination & backup directories for the pictures
dst = "pic" #destination directory for images
bkp = "pic_bkp" #backup  directory for images

#if the picture directories don't exist, create them
if not os.path.exists("pic"):
	os.makedirs("pic")
if not os.path.exists("pic_bkp"):
	os.makedirs("pic_bkp")

#create a loop that constantly grabs new images from the webcam
while True:
        #set a time variable that updates with the loop
        current_time = time.time()
        #grab an image still from the camera and convert it to grayscale
        img01 = cam.getImage().toGray()
        #wait half a second
        time.sleep(0.5)
	#grab an unedited still to use as our original image
	original = cam.getImage()
        #grab another image still from the camera and conver it to grayscale
        img02 = cam.getImage().toGray()
        #subract the images from each other, binarize and inver the colors
        diff = (img01 - img02).binarize(50).invert()

        #dump all the values into a Numpy matrix and extract the mean avg
        matrix = diff.getNumpy()
        mean = matrix.mean()

	#find and highlight the objects within the image
	blobs = diff.findBlobs()

        #check to see if the wait time has been passed
	if current_time >= (start_time + wait_time):
		#if it has, reset the start time
		start_time = time.time()
		#scan the picture directory for files
		for root, dirs, files in os.walk(dst):
			dst_root = root.replace(dst, bkp)
			#if a file is found in the picture directory, send it to email
			if files:
				firstfile = sorted(files)[0]
				img_mailer = os.path.join(root, firstfile)
				py_gmailer.gmail(img_mailer)
			#move any files in the pic directory to the backup directory
			for file_ in files:
				src_file = os.path.join(root, file_)
				dst_file = os.path.join(dst_root, file_)
				shutil.move(src_file, dst_root)

        #if the mean is greater than our threshold variable, then look for objects
	if mean >= threshold:

		#check to see if any objects were detected
		if blobs:
			#find the central point of each object
			#and draw a red circle around it
			for b in blobs:
				try:
					loc = (b.x,b.y) #locates center of object
					original.drawCircle(loc,b.radius(),Color.RED,2)
				except:
					e = sys.exc_info()[0]
		#use the current date to create a unique file name
		timestr = time.strftime("%Y%m%d-%H%M%S")
		
		#initialize the counter variable
		i = 1
		
		#check to see if the filename already exists
		while os.path.exists("pic/motion%s-%s.png" % (timestr, i)):
			#if it does, add one to the filename and try again
			i += 1
		#once a unique filename has been found, save the image
		original.save("pic/motion%s-%s.png" % (timestr, i))
		#print results to terminal
		print("Motion Detected")

	#send the current image to the webcam stream
	original.save(streaming)