Video recognition and Web API learning

trackSpecies.py

import sys
import time  
import httplib, urllib
import cv2
import serial

from poster.encode import multipart_encode
from poster.streaminghttp import register_openers
import urllib2
import json

thingSpeakApiKey = "O93NMWVP2BW5NQME" 

# Load the XML for trained information
cascPath = '/usr/local/share/OpenCV/haarcascades/haarcascade_frontalface_alt.xml'
faceCascade = cv2.CascadeClassifier(cascPath)

# Set /dev/video0 as device
video_capture = cv2.VideoCapture(0)

# Set serial port for Arduino Uno
ser = serial.Serial( "/dev/ttyUSB0", baudrate=9600, timeout=0.5 )

def post_to_thingspeak(payload):  
    headers = {"Content-type": "application/x-www-form-urlencoded","Accept": "text/plain"}
    not_connected = 1
    while (not_connected):
        try:
            conn = httplib.HTTPConnection("api.thingspeak.com:80")
            conn.connect()
            not_connected = 0
        except (httplib.HTTPException, socket.error) as ex:
            print "Error: %s" % ex
            # time.sleep(10)
    conn.request("POST", "/update", payload, headers)
    response = conn.getresponse()
    print( response.status, response.reason, payload, time.strftime("%c"))
    data = response.read()
    conn.close()

def learning():
    register_openers()
    datagen, headers = multipart_encode({"pic": open("/tmp/temp.jpg")})
    upload_url = 'http://scsonic.com/nasa/upload.php'
    request = urllib2.Request(upload_url, datagen, headers)
    j = urllib2.urlopen(request).read()
    dj = json.loads(j)
    return dj["catch"] 

while True:
    try:
        # Capture frame-by-frame
        ret, frame = video_capture.read()

        gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

        faces = faceCascade.detectMultiScale(
            gray,
            scaleFactor=1.1,
            minNeighbors=5,
            minSize=(30, 30),
            flags=cv2.cv.CV_HAAR_SCALE_IMAGE
        )

        # Draw a rectangle around the faces
        for (x, y, w, h) in faces:
            cv2.rectangle(frame, (x, y), (x+w, y+h), (0, 255, 0), 2)

        # Display the resulting frame
        cv2.imshow('Video', frame)

        # Write a copy to memory to learn by TensorFlow
        cv2.imwrite('/tmp/temp.jpg', frame)

        if learning() == True:
           # Query the environemtal informations and catch it!!!
           ser.write("B")
           env = ser.readline()
           e = env.split()
           params = urllib.urlencode({'field1': e[0], 'field2': e[1], 'field3': e[2], 'field4': e[3], 'field5': e[4], 'key': thingSpeakApiKey})
           post_to_thingspeak(params)
        else:
           ser.write("R")

        if cv2.waitKey(1) & 0xFF == ord('q'):
            break
    except:
      pass

# When everything is done, release the capture
video_capture.release()
cv2.destroyAllWindows()

# Close the serial port
ser.close()