Project_ML_RPi_Code

#!/usr/bin/python3
import subprocess
import numpy as np
import re
import cv2
from tflite_runtime.interpreter import Interpreter
import os
import time
import sys
import math
import glob
import signal
from datetime import datetime,timedelta
#import pygame
from pygame import mixer
#pygame.mixer.init()
mixer.init()

def alert(audio, timer):
    mixer.music.load(audio)
    mixer.music.set_volume(0.7)
    mixer.music.play()
    #pygame.mixer.Sound(audio).play()
    time.sleep(timer)
    #pygame.mixer.Sound(audio).stop()
    #mixer.music.stop()
    time.sleep(0.5)
    print("Audio File Finished Playing")

check = 1

# clear ram
pics = glob.glob('/run/shm/test*.jpg')
for t in range(0,len(pics)):
    os.remove(pics[t])

def Camera_start(wx,hx):
    global p
    rpistr = "libcamera-vid -t 0 --segment 1 --codec mjpeg -n -o /run/shm/test%06d.jpg --width " + str(wx) + " --height " + str(hx)
    p = subprocess.Popen(rpistr, shell=True, preexec_fn=os.setsid)

#initialise variables
width        = 640 #720
height       = 480 #540
CAMERA_WIDTH = 640
CAMERA_HEIGHT = 480
#face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
#eye_cascade  = cv2.CascadeClassifier('haarcascade_eye.xml')
#open_eyes_detected = 0
#face_detected = 0
start = 0
cv2.namedWindow('Frame')
Text = "Left Mouse click on picture to EXIT, Right Mouse click for eye detaction ON/OFF"
ttrat = time.time()

#define mouse clicks (LEFT to EXIT, RIGHT to switch eye detcetion ON/OFF)
def mouse_action(event, x, y, flags, param):
    global p,check
    if event == cv2.EVENT_LBUTTONDOWN:
        os.killpg(p.pid, signal.SIGTERM)
        cv2.destroyAllWindows()
        sys.exit()
    if event == cv2.EVENT_RBUTTONDOWN:
        if check == 0:
            check = 1
        else:
            check = 0

cv2.setMouseCallback('Frame',mouse_action)

# start capturing images
Camera_start(width,height)

def load_labels(path='labels.txt'):
  """Loads the labels file. Supports files with or without index numbers."""
  with open(path, 'r', encoding='utf-8') as f:
    lines = f.readlines()
    labels = {}
    for row_number, content in enumerate(lines):
      pair = re.split(r'[:\s]+', content.strip(), maxsplit=1)
      if len(pair) == 2 and pair[0].strip().isdigit():
        labels[int(pair[0])] = pair[1].strip()
      else:
        labels[row_number] = pair[0].strip()
  return labels

def set_input_tensor(interpreter, image):
  """Sets the input tensor."""
  tensor_index = interpreter.get_input_details()[0]['index']
  input_tensor = interpreter.tensor(tensor_index)()[0]
  input_tensor[:, :] = np.expand_dims((image-255)/255, axis=0)


def get_output_tensor(interpreter, index):
  """Returns the output tensor at the given index."""
  output_details = interpreter.get_output_details()[index]
  tensor = np.squeeze(interpreter.get_tensor(output_details['index']))
  return tensor


def detect_objects(interpreter, image, threshold):
  """Returns a list of detection results, each a dictionary of object info."""
  set_input_tensor(interpreter, image)
  interpreter.invoke()
  # Get all output details
  boxes = get_output_tensor(interpreter, 1)
  classes = get_output_tensor(interpreter, 3)
  scores = get_output_tensor(interpreter, 0)
  count = int(get_output_tensor(interpreter, 2))

  results = []
  for i in range(count):
    if scores[i] >= threshold:
      result = {
          'bounding_box': boxes[i],
          'class_id': classes[i],
          'score': scores[i]
      }
      results.append(result)
  return results

def main():
    labels = load_labels()
    interpreter = Interpreter('detect.tflite')
    interpreter.allocate_tensors()
    _, input_height, input_width, _ = interpreter.get_input_details()[0]['shape']

    #cap = cv2.VideoCapture(0)
    while True:

            # remove message after 3 seconds
        #if time.time() - ttrat > 3 and ttrat > 0:
        #    Text =""
        #    ttrat = 0

    # load image
        pics = glob.glob('/run/shm/test*.jpg')
        while len(pics) < 2:
            pics = glob.glob('/run/shm/test*.jpg')
        pics.sort(reverse=True)
        imgf = cv2.imread(pics[1])
        if len(pics) > 2:
            for tt in range(2,len(pics)):
                os.remove(pics[tt])

        #ret, frame = cap.read()
        img = cv2.resize(cv2.cvtColor(imgf, cv2.COLOR_BGR2RGB), (320,320))
        res = detect_objects(interpreter, img, 0.7)
        print(res)


        for result in res:
            ymin, xmin, ymax, xmax = result['bounding_box']
            xmin = int(max(1,xmin * CAMERA_WIDTH))
            xmax = int(min(CAMERA_WIDTH, xmax * CAMERA_WIDTH))
            ymin = int(max(1, ymin * CAMERA_HEIGHT))
            ymax = int(min(CAMERA_HEIGHT, ymax * CAMERA_HEIGHT))

            cv2.rectangle(imgf,(xmin, ymin),(xmax, ymax),(0,255,0),3)
            cv2.putText(imgf,labels[int(result['class_id'])],(xmin, min(ymax, CAMERA_HEIGHT-20)), cv2.FONT_HERSHEY_SIMPLEX, 0.5,(255,255,255),2,cv2.LINE_AA)

            # Website to Convert from Whatsapp voice to [mp3] - https://convertio.co/ogg-mp3/
            if labels[int(result['class_id'])] == 'goodjob':
                print('play SOUND Good Job')
                alert("goodjob.mp3",3)
            if labels[int(result['class_id'])] == 'goodlife':
                print('play SOUND Good Life')
                alert("goodlife.mp3",3)
            if labels[int(result['class_id'])] == 'hello':
                print('play SOUND HELLO')
                alert("hello.mp3",3)
            if labels[int(result['class_id'])] == 'iloveyou':
                print('play SOUND I Love You')
                alert("iloveyou.mp3",3)
            if labels[int(result['class_id'])] == 'notsure':
                print('play SOUND Not Sure')
                alert("notsure.mp3",3)
            if labels[int(result['class_id'])] == 'ok':
                print('play SOUND OK')
                alert("ok.mp3",3)
            if labels[int(result['class_id'])] == 'thisiscool':
                print('play SOUND This Is Cool')
                alert("thisiscool.mp3",3)
        #cv2.imshow('Pi Feed', frame)
        #display image
        cv2.putText(img,Text, (10, height - 10), 0, 0.4, (0, 255, 255))
        cv2.imshow('Frame',imgf)
        cv2.waitKey(10)

        if cv2.waitKey(10) & 0xFF ==ord('q'):
            cap.release()
            cv2.destroyAllWindows()

if __name__ == "__main__":
    main()