import streamlit as st
import cv2
import numpy as np
from deepface import DeepFace
import os
import io
from PIL import Image
import tensorflow as tf

# Suppress TensorFlow and oneDNN warnings
os.environ["TF_ENABLE_ONEDNN_OPTS"] = "0"
tf.get_logger().setLevel('ERROR') # Suppresses most TensorFlow logs

# Set Streamlit page config for theme
st.set_page_config(page_title="Real-Time Emotion Detection", page_icon="😊", layout="wide")

# Custom CSS for background styling
st.markdown(
"""
<style>
body {
background-color: #f0f2f6;
}
.stApp {
background: linear-gradient(to right,rgb(95, 159, 255),rgb(123, 223, 254));
color: white;
}
</style>
""",
unsafe_allow_html=True
)

st.title("😊 Real-Time Emotion Detection")

# Initialize webcam using OpenCV
cap = cv2.VideoCapture(0)
if not cap.isOpened():
st.error("Error: Could not open webcam.")
cap.release()
st.stop()

face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + "haarcascade_frontalface_default.xml")
frame_placeholder = st.empty()
stop_button = st.button("Stop Camera", key="stop_button")

while cap.isOpened():
ret, frame = cap.read()
if not ret:
st.error("Failed to capture image")
break

gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, scaleFactor=1.3, minNeighbors=5, minSize=(50, 50))

for (x, y, w, h) in faces:
face = frame[y:y + h, x:x + w]
try:
result = DeepFace.analyze(face, actions=['emotion'], enforce_detection=False)
emotion = result[0]['dominant_emotion'] if isinstance(result, list) else result['dominant_emotion']
cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 2)
cv2.putText(frame, f"Emotion: {emotion}", (x, y - 10),
cv2.FONT_HERSHEY_SIMPLEX, 0.8, (0, 255, 255), 2, cv2.LINE_AA)
except Exception as e:
st.error(f"Error detecting emotion: {e}")

# Convert to PIL format for display
frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
image = Image.fromarray(frame)
frame_placeholder.image(image, caption="Live Emotion Detection", use_container_width=True)

if stop_button:
break

cap.release()
st.write("✅ Camera released. Application terminated.")