Face Detection Using Python in Android With Pydroid Easy Step By Step

#1 Installing Requirements from play store

Search and Install 

1) Pydroid permissions plugin 

2) Pydroid repository plugin

3) Pydroid 3 - IDE for  Python 3

#2 Installing Modues from Pydroid 

Now open Pydroid 3 App after open of Pydroid 3 wait a minutes to install python

Now time to install module

But first note you have to buy Get Premium in Just ₹65 then you can able to install opencv-python module.

After Buy premium

Now Click On PIP

Then click on Quick Install now install opencv-python and PyQt5 

#3 Write Code And Run It

Now Write Code

import cv2

import numpy as np

import sys

from PyQt5 import QtCore

from PyQt5 import QtWidgets

from PyQt5 import QtGui

from PyQt5.QtGui import QImage

class RecordVideo(QtCore.QObject):

    image_data = QtCore.pyqtSignal(np.ndarray)

    def __init__(self, camera_port, parent=None):

        super().__init__(parent)

        self.camera = cv2.VideoCapture(camera_port)

        self.camera.set(cv2.CAP_PROP_FRAME_WIDTH, 640)

        self.camera.set(cv2.CAP_PROP_FRAME_HEIGHT, 480)

        self.timer = QtCore.QTimer()

        self.timer.timeout.connect(self.timerEvent)

        self.timer.start(0)

    def timerEvent(self):

        read, data = self.camera.read()

        if read:

            self.image_data.emit(data)

class FaceDetectionWidget(QtWidgets.QWidget):

    def __init__(self, cascade_filepath, parent=None):

        super().__init__(parent)

        self.classifier = cv2.CascadeClassifier(cascade_filepath)

        self.image = QImage()

        self._border = (0, 255, 0)

        self._width = 2

    def detect_faces(self, image):

        gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

        gray_image = cv2.equalizeHist(gray_image)

        faces = self.classifier.detectMultiScale(gray_image, 1.3, 5)

        return faces

    def image_data_slot(self, image_data):

        if (self.width() > self.height()) != (image_data.shape[1] > image_data.shape[0]):

            # Need to rotate image data, the screen / camera is rotated

            image_data = cv2.rotate(image_data, cv2.ROTATE_90_COUNTERCLOCKWISE)

        faces = self.detect_faces(image_data)

        for (x, y, w, h) in faces:

            cv2.rectangle(image_data, (x, y), (x + w, y + h), self._border, self._width)

        self.image = self.get_qimage(image_data)

        self.update()

    def get_qimage(self, image):

        height, width, colors = image.shape

        image = QImage(image.data, width, height, 3 * width, QImage.Format_RGB888).rgbSwapped()

        return image

    def paintEvent(self, event):

        painter = QtGui.QPainter(self)

        w = self.width()

        h = self.height()

        cw = self.image.width()

        ch = self.image.height()

        # Keep aspect ratio

        if ch != 0 and cw != 0:

            w = min(cw * h / ch, w)

            h = min(ch * w / cw, h)

            w, h = int(w), int(h)

        painter.drawImage(QtCore.QRect(0, 0, w, h), self.image)

        self.image = QImage()

class MainWidget(QtWidgets.QWidget):

    def __init__(self, haarcascade_filepath, parent=None):

        super().__init__(parent)

        fp = haarcascade_filepath

        self.face_detection_widget = FaceDetectionWidget(fp)

        # 1 is used for frontal camera

        self.record_video = RecordVideo(1)

        self.record_video.image_data.connect(self.face_detection_widget.image_data_slot)

        layout = QtWidgets.QVBoxLayout()

        layout.addWidget(self.face_detection_widget)

        self.setLayout(layout)

app = QtWidgets.QApplication(sys.argv)

haar_cascade_filepath = cv2.data.haarcascades + '/haarcascade_frontalface_default.xml'

main_window = QtWidgets.QMainWindow()

main_widget = MainWidget(haar_cascade_filepath)

main_window.setCentralWidget(main_widget)

main_window.show()

sys.exit(app.exec_())

Let's Run the code

If you have any problem then contact me on telegram @odewithyash