Train.py

from keras.models import Sequential
from keras.layers import Convolution2D
from keras.layers import MaxPooling2D
from keras.layers import Flatten
from keras.layers import Dense
import cv2
import numpy as np
import os

cap = cv2.VideoCapture(0)

while True:
    _, frame = cap.read()
 
    frame = cv2.flip(frame, 1)
    
    x1 = int(0.5*frame.shape[1])
    y1 = 10
    x2 = frame.shape[1]-10
    y2 = int(0.5*frame.shape[1])
    cv2.rectangle(frame, (x1-1, y1-1), (x2+1, y2+1), (255,0,0) ,1)

    roi = frame[y1:y2, x1:x2]
    roi = cv2.resize(roi, (64, 64)) 
 
    cv2.imshow("Frame", frame)

    roi = cv2.cvtColor(roi, cv2.COLOR_BGR2GRAY)
    _, roi = cv2.threshold(roi, 120, 255, cv2.THRESH_BINARY)
    cv2.imshow("ROI", roi)
    
    interrupt = cv2.waitKey(10)

cap.release()
cv2.destroyAllWindows()

classifier = Sequential()

classifier.add(Convolution2D(32, (3, 3), input_shape=(64, 64, 1), activation='relu'))
classifier.add(MaxPooling2D(pool_size=(2, 2)))

classifier.add(Convolution2D(32, (3, 3), activation='relu'))

classifier.add(MaxPooling2D(pool_size=(2, 2)))

classifier.add(Flatten())

classifier.add(Dense(units=128, activation='relu'))
classifier.add(Dense(units=6, activation='softmax')) 

classifier.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy']) 

from keras.preprocessing.image import ImageDataGenerator

traindata = ImageDataGenerator(
        rescale=1./255,
        shear_range=0.2,
        zoom_range=0.2,
        horizontal_flip=True)

testdata = ImageDataGenerator(rescale=1./255)

training_set = traindata.flow_from_directory('data/train',
                                                 target_size=(64, 64),
                                                 batch_size=5,
                                                 color_mode='grayscale',
                                                 class_mode='categorical')

test_set = testdata.flow_from_directory('data/test',
                                            target_size=(64, 64),
                                            batch_size=5,
                                            color_mode='grayscale',
                                            class_mode='categorical') 
classifier.fit_generator(
        training_set,
        steps_per_epoch=600, 
        epochs=10,
        validation_data=test_set,
        validation_steps=30)

model_save = classifier