本网页已闲置超过3分钟,按键盘任意键或点击空白处,即可回到网页

基于树莓派的快速叫醒装置
发布时间:2021-05-17
分享到:
基于树莓派的快速叫醒装置
发布时间:2021-05-17
分享到:

在过去的几年里,我每天早上都难以醒来。我试过很多方法,比如提醒自己起床,强迫自己起床,但是没有一个奏效。我决定想出一个解决办法来解决这些问题,如果闹钟响后5分钟我还没起床,这个装置就会把枕头扔到我头上。

该项目所设计的电子设备由树莓派3型、树莓派摄像头、3根公母跳线、微型伺服电机和电池组组成。

硬件组件:

  • 树莓派 3 型
  • 树莓派摄像机
  • 跳线
  • 微型伺服电机
  • 电池

示意图:

代码:

该电子设备的代码由4部分构成。

1)face_shot.py的文件。它用于拍摄您的面部并收集数据,然后训练模型。

import cv2

name = 'will'

cam = cv2.VideoCapture(0)

cv2.namedWindow("press space to take a photo", cv2.WINDOW_NORMAL)
cv2.resizeWindow("press space to take a photo", 500, 300)

img_counter = 0

while True:
    ret, frame = cam.read()
    if not ret:
        print("failed to grab frame")
        break
    cv2.imshow("press space to take a photo", frame)

    k = cv2.waitKey(1)
    if k%256 == 27:
        # ESC pressed
        print("Escape hit, closing...")
        break
    elif k%256 == 32:
        # SPACE pressed
        img_name = "dataset/"+ name +"/image_{}.jpg".format(img_counter)
        cv2.imwrite(img_name, frame)
        print("{} written!".format(img_name))
        img_counter += 1

cam.release()

cv2.destroyAllWindows()

2)train_model.py的文件。它用于根据您使用face_shot.py拍摄的图像来训练模型。

#! /usr/bin/python

# import the necessary packages
from imutils import paths
import face_recognition
#import argparse
import pickle
import cv2
import os

# our images are located in the dataset folder
print("[INFO] start processing faces...")
imagePaths = list(paths.list_images("dataset"))

# initialize the list of known encodings and known names
knownEncodings = []
knownNames = []

# loop over the image paths
for (i, imagePath) in enumerate(imagePaths):
    # extract the person name from the image path
    print("[INFO] processing image {}/{}".format(i + 1,
        len(imagePaths)))
    name = imagePath.split(os.path.sep)[-2]

    # load the input image and convert it from RGB (OpenCV ordering)
    # to dlib ordering (RGB)
    image = cv2.imread(imagePath)
    rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

    # detect the (x, y)-coordinates of the bounding boxes
    # corresponding to each face in the input image
    boxes = face_recognition.face_locations(rgb,
        model="hog")

    # compute the facial embedding for the face
    encodings = face_recognition.face_encodings(rgb, boxes)

    # loop over the encodings
    for encoding in encodings:
        # add each encoding + name to our set of known names and
        # encodings
        knownEncodings.append(encoding)
        knownNames.append(name)

# dump the facial encodings + names to disk
print("[INFO] serializing encodings...")
data = {"encodings": knownEncodings, "names": knownNames}
f = open("encodings.pickle", "wb")
f.write(pickle.dumps(data))
f.close()

3)face_rec.py的文件。它是您想要实际启动面部识别软件时运行的文件。

#! /usr/bin/python

# import the necessary packages
from datetime import datetime
import servo_move
from imutils.video import VideoStream
from imutils.video import FPS
import face_recognition
import imutils
import pickle
import time
import cv2

now = datetime.now()
da_time = datetime(2021, 4, 7, 12, 35, 00)
x = 0
#Initialize 'currentname' to trigger only when a new person is identified.
currentname = "unknown"
#Determine faces from encodings.pickle file model created from train_model.py
encodingsP = "encodings.pickle"
#use this xml file
#https://github.com/opencv/opencv/blob/master/data/haarcascades/haarcascade_frontalface_default.xml
cascade = "haarcascade_frontalface_default.xml"

# load the known faces and embeddings along with OpenCV's Haar
# cascade for face detection
print("[INFO] loading encodings + face detector…")
data = pickle.loads(open(encodingsP, "rb").read())
detector = cv2.CascadeClassifier(cascade)

# initialize the video stream and allow the camera sensor to warm up
print("[INFO] starting video stream…")
vs = VideoStream(src=0).start()
#vs = VideoStream(usePiCamera=True).start()
time.sleep(2.0)

# start the FPS counter
fps = FPS().start()

# loop over frames from the video file stream
while True:
    # grab the frame from the threaded video stream and resize it
    # to 500px (to speedup processing)
    frame = vs.read()
    frame = imutils.resize(frame, width=500)
    
    # convert the input frame from (1) BGR to grayscale (for face
    # detection) and (2) from BGR to RGB (for face recognition)
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)

    # detect faces in the grayscale frame
    rects = detector.detectMultiScale(gray, scaleFactor=1.1, 
        minNeighbors=5, minSize=(30, 30),
        flags=cv2.CASCADE_SCALE_IMAGE)

    # OpenCV returns bounding box coordinates in (x, y, w, h) order
    # but we need them in (top, right, bottom, left) order, so we
    # need to do a bit of reordering
    boxes = [(y, x + w, y + h, x) for (x, y, w, h) in rects]

    # compute the facial embeddings for each face bounding box
    encodings = face_recognition.face_encodings(rgb, boxes)
    names = []

    # loop over the facial embeddings
    for encoding in encodings:
        # attempt to match each face in the input image to our known
        # encodings
        matches = face_recognition.compare_faces(data["encodings"],
            encoding)
        name = "Unknown" #if face is not recognized, then print Unknown

        # check to see if we have found a match
        if True in matches:
            # find the indexes of all matched faces then initialize a
            # dictionary to count the total number of times each face
            # was matched
            matchedIdxs = [i for (i, b) in enumerate(matches) if b]
            counts = {}

            # loop over the matched indexes and maintain a count for
            # each recognized face face
            for i in matchedIdxs:
                name = data["names"][i]
                counts[name] = counts.get(name, 0) + 1

            # determine the recognized face with the largest number
            # of votes (note: in the event of an unlikely tie Python
            # will select first entry in the dictionary)
            name = max(counts, key=counts.get)
            
            #If someone in your dataset is identified, print their name on the screen
            if currentname != name:
                currentname = name
                print(currentname)
        
        # update the list of names
        names.append(name)


    # loop over the recognized faces
    for ((top, right, bottom, left), name) in zip(boxes, names):
        # draw the predicted face name on the image – color is in BGR
        cv2.rectangle(frame, (left, top), (right, bottom),
            (0, 255, 0), 2)
        y = top - 15 if top - 15 > 15 else top + 15
        cv2.putText(frame, name, (left, y), cv2.FONT_HERSHEY_SIMPLEX,
            .8, (255, 0, 0), 2)


    # display the image to our screen
    cv2.imshow("Facial Recognition is Running", frame)
    key = cv2.waitKey(1) & 0xFF

    # quit when 'q' key is pressed
    if key == ord("q"):
        break

    # update the FPS counter
    fps.update()

    current_time = datetime.now()
    if (currentname == "will") and (current_time.time() > da_time.time()) and (x == 0):
        exec(open("servo_move.py").read())
        x = 1

# stop the timer and display FPS information
fps.stop()
print("[INFO] elasped time: {:.2f}".format(fps.elapsed()))
print("[INFO] approx. FPS: {:.2f}".format(fps.fps()))

# do a bit of cleanup
cv2.destroyAllWindows()
vs.stop()

4) servo_move.py的文件。伺服移动,它使伺服移动180度,然后向后移动。

import RPi.GPIO as GPIO
import time

GPIO.setmode(GPIO.BOARD)

GPIO.setup(11,GPIO.OUT)
servo1 = GPIO.PWM(11,50)
servo1.start(0)
servo1.ChangeDutyCycle(12)
time.sleep(2)
servo1.ChangeDutyCycle(2)
time.sleep(0.5)
servo1.ChangeDutyCycle(0)
servo1.stop()
GPIO.cleanup()

制造:

第一部分:

将伺服引脚连接到树莓pi上的引脚4、6和11。我还把相机插在树莓皮上带状电缆的小插槽里。

第二部分:

有3D打印的木头与一些铰链连接。

加入微信技术交流群

技术交流,职业进阶

关注与非网服务号

获取电子工程师福利

加入电路城 QQ 交流群

与技术大牛交朋友

讨论