import face_recognition
import os
import cv2
import numpy as np
import time
from keras.models import load_model
from PIL import Image
import streamlit as st

# Streamlit UI Setup
st.markdown("<h1 style='text-align: center;'>Emotion & Face Recognition</h1>", unsafe_allow_html=True)
st.markdown("<h3 style='text-align: center;'>angry, fear, happy, neutral, sad, surprise</h3>", unsafe_allow_html=True)

# Known faces folder path
KNOWN_FACES_DIR = "known_faces"

# Load emotion detection model
@st.cache_resource
def load_emotion_model():
    return load_model("CNN_Model_acc_75.h5")

emotion_model = load_emotion_model()

# Face detection model
face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + "haarcascade_frontalface_default.xml")
emotion_labels = ['angry', 'fear', 'happy', 'neutral', 'sad', 'surprise']
img_shape = 48

# Known faces dictionary
known_faces = {"names": [], "encodings": []}

def load_faces_from_folder(folder_path):
    """
    Load known faces from a folder, using filenames as names.
    """
    for filename in os.listdir(folder_path):
        if filename.endswith(('.jpg', '.jpeg', '.png')):
            name = os.path.splitext(filename)[0]
            image_path = os.path.join(folder_path, filename)
            
            # Load and encode the image
            image = face_recognition.load_image_file(image_path)
            face_encodings = face_recognition.face_encodings(image)
            
            if face_encodings:  # Ensure a face is found
                known_faces["names"].append(name)
                known_faces["encodings"].append(face_encodings[0])
                print(f"Loaded face for {name}")
            else:
                print(f"No face detected in {filename}")

# Load known faces
load_faces_from_folder(KNOWN_FACES_DIR)

def recognize_face(unknown_face_encoding):
    """
    Compare an unknown face with the known faces and return the closest match.
    """
    matches = face_recognition.compare_faces(known_faces["encodings"], unknown_face_encoding, tolerance=0.6)
    if True in matches:
        match_index = matches.index(True)
        return known_faces["names"][match_index]
    return "Unknown"

def detect_emotion(face_image):
    """
    Predict the emotion of a face using the emotion detection model.
    """
    face_resized = cv2.resize(face_image, (img_shape, img_shape))
    face_resized = np.expand_dims(face_resized, axis=0)
    face_resized = face_resized / 255.0  # Normalize the image
    predictions = emotion_model.predict(face_resized)
    return emotion_labels[np.argmax(predictions)]

def process_frame_with_recognition_and_emotion(frame):
    """
    Detect faces, recognize names, and detect emotions in the frame.
    """
    gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
    rgb_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
    faces = face_cascade.detectMultiScale(gray_frame, scaleFactor=1.1, minNeighbors=5, minSize=(30, 30))

    for (x, y, w, h) in faces:
        # Get the face area
        face_image = rgb_frame[y:y+h, x:x+w]
        face_encodings = face_recognition.face_encodings(face_image)

        if face_encodings:
            name = recognize_face(face_encodings[0])  # Recognize the face
        else:
            name = "Unknown"

        # Predict emotion
        emotion = detect_emotion(frame[y:y+h, x:x+w])

        # Display name and emotion
        display_text = f"{name} is Feeling {emotion}"
        cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
        cv2.putText(frame, display_text, (x, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.8, (0, 255, 0), 2)

    return frame

def video_feed(video_source):
    """
    Display video feed with face recognition and emotion detection.
    """
    frame_placeholder = st.empty()  # Placeholder for updating frames

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

        frame = process_frame_with_recognition_and_emotion(frame)
        frame_placeholder.image(frame, channels="BGR", use_column_width=True)

# Sidebar options
upload_choice = st.sidebar.radio("Choose input source", ["Upload Image", "Upload Video", "Camera"])

if upload_choice == "Camera":
    video_source = cv2.VideoCapture(0)  # Access webcam
    video_feed(video_source)

elif upload_choice == "Upload Video":
    uploaded_video = st.file_uploader("Upload Video", type=["mp4", "mov", "avi", "mkv", "webm"])
    if uploaded_video:
        with tempfile.NamedTemporaryFile(delete=False) as tfile:
            tfile.write(uploaded_video.read())
            video_source = cv2.VideoCapture(tfile.name)
            video_feed(video_source)

elif upload_choice == "Upload Image":
    uploaded_image = st.file_uploader("Upload Image", type=["png", "jpg", "jpeg"])
    if uploaded_image:
        image = Image.open(uploaded_image)
        frame = np.array(image)
        frame = process_frame_with_recognition_and_emotion(frame)
        st.image(frame, caption="Processed Image", use_column_width=True)