new.py

# 


import cv2
import numpy as np
import tensorflow as tf

# Load the face detector with the correct file path
face_detector = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
# Load the emotion model
emotion_dict = {0: "Angry", 1: "Disgusted", 2: "Fearful", 3: "Happy", 4: "Neutral", 5: "Sad", 6: "Surprised"}
emotion_model = tf.keras.models.load_model("model_emotion.h5")
emotion_model.load_weights("model_weights_new.h5")
print("Loaded emotion model from disk")

# Define the predict_img function
def predict_img(frame):
    # Resize the image
    frame = cv2.resize(frame, (1280, 720))
    num_faces = face_detector.detectMultiScale(frame, scaleFactor=1.3, minNeighbors=5)

    # Draw bounding boxes and annotate the image
    for (x, y, w, h) in num_faces:
        cv2.rectangle(frame, (x, y-50), (x+w, y+h+10), (0, 255, 0), 4)
        roi_gray_frame = frame[y:y + h, x:x + w]

        # Preprocess the input image
        resized_img = cv2.resize(roi_gray_frame, (48, 48))
        gray_img = cv2.cvtColor(resized_img, cv2.COLOR_BGR2GRAY)
        input_img = np.expand_dims(gray_img, axis=-1)  # Add the channel dimension
        input_img = np.expand_dims(input_img, axis=0)  # Add the batch dimension

        # Normalize the image
        input_img = input_img / 255.0

        # Predict the emotions
        emotion_prediction = emotion_model.predict(input_img)
        maxindex = int(np.argmax(emotion_prediction))
        emotion_label = emotion_dict[maxindex]

        # Annotate the image with emotion label
        cv2.putText(frame, emotion_label, (x+5, y-20), cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 0, 0), 2, cv2.LINE_AA)

    return frame

# Capture video from webcam
cap = cv2.VideoCapture(0)

while True:
    ret, frame = cap.read()
    if not ret:
        break

    annotated_frame = predict_img(frame)
    cv2.imshow('Emotion Detection', annotated_frame)

    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

cap.release()
cv2.destroyAllWindows()