import numpy as np
import cv2
from tensorflow.keras.models import model_from_json

model_json_file = './input/model.json'
model_weights_file = './input/model_weights.h5'
with open(model_json_file, "r") as json_file:
    loaded_model_json = json_file.read()
    loaded_model = model_from_json(loaded_model_json)
    loaded_model.load_weights(model_weights_file)

face_cascade = cv2.CascadeClassifier('./input/haarcascade_frontalface_default.xml')
img_size = 64
cap = cv2.VideoCapture(0)
import copy

while True:
    
    ret, frame = cap.read()
    img = copy.deepcopy(frame)
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    faces = face_cascade.detectMultiScale(gray, 1.3, 5)
    for (x,y,w,h) in faces:
        fc = gray[y:y+h, x:x+w]
        
        roi = cv2.resize(fc, (img_size,img_size))
        pred = loaded_model.predict(roi[np.newaxis, :, :, np.newaxis])
        text_idx=np.argmax(pred)
        text_list = ['Angry', 'Disgust', 'Fear', 'Happy', 'Neutral', 'Sad', 'Surprise']
        if text_idx == 0:
            text= text_list[0]
        if text_idx == 1:
            text= text_list[1]
        elif text_idx == 2:
            text= text_list[2]
        elif text_idx == 3:
            text= text_list[3]
        elif text_idx == 4:
            text= text_list[4]
        elif text_idx == 5:
            text= text_list[5]
        elif text_idx == 6:
            text= text_list[6]
        cv2.putText(img, text, (x, y-5),
           cv2.FONT_HERSHEY_SIMPLEX, 0.45, (255, 0, 255), 2)
        img = cv2.rectangle(img, (x,y), (x+w, y+h), (0,0,255), 2)
            
    
    cv2.imshow("frame", img)
    key = cv2.waitKey(1) & 0xFF
    if key== ord('q'):
        break
    
cap.release()
cv2.destroyAllWindows()