app.py

import streamlit as st
import numpy as np
import cv2
from keras.models import load_model
from keras.preprocessing.image import img_to_array
from PIL import Image
import base64

# Set page config
st.set_page_config(page_title="Face Mask Detection", layout="centered")

# Function to set background image
def set_background(image_path):
    with open(image_path, "rb") as img_file:
        encoded = base64.b64encode(img_file.read()).decode()
    st.markdown(
        f"""
        <style>
        .stApp {{
            background-image: url("data:image/jpg;base64,{encoded}");
            background-size: cover;
            background-position: center;
            background-repeat: no-repeat;
        }}
        </style>
        """,
        unsafe_allow_html=True
    )

# Set background image
set_background("IMAGE_1.jpg")

# Load model once
@st.cache_resource
def load_model_cached():
    return load_model("FaceDetection.keras")  # Ensure model is trained on cropped face images

model = load_model_cached()

# Load Haar Cascade for face detection
face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + "haarcascade_frontalface_default.xml")

# Function to detect face and predict
def detect_and_predict(image_input):
    image_np = np.array(image_input.convert("RGB"))
    gray = cv2.cvtColor(image_np, cv2.COLOR_RGB2GRAY)

    faces = face_cascade.detectMultiScale(gray, 1.1, 4)

    if len(faces) == 0:
        return image_input, None, "No face detected"

    x, y, w, h = faces[0]
    face_roi = image_np[y:y+h, x:x+w]
    face_pil = Image.fromarray(face_roi).resize((200, 200))
    img_array = img_to_array(face_pil) / 255.0
    img_array = np.expand_dims(img_array, axis=0)

    prediction = model.predict(img_array)[0][0]
    confidence = (1 - prediction) if prediction < 0.5 else prediction
    label = "✅ Mask Detected" if prediction < 0.5 else "🚫 No Mask Detected"

    # Draw rectangle and label
    color = (0, 255, 0) if prediction < 0.5 else (255, 0, 0)
    cv2.rectangle(image_np, (x, y), (x + w, y + h), color, 2)
    cv2.putText(image_np, f"{label} ({confidence*100:.2f}%)", (x, y - 10),
                cv2.FONT_HERSHEY_SIMPLEX, 0.4, color, 2)

    return Image.fromarray(image_np), confidence, label

# App UI
st.title("Face Mask Detection")
st.markdown("Upload a face image or use your webcam to check if a mask is being worn.")

# Tabs
tab1, tab2 = st.tabs([" Upload Image", "Use Webcam"])

with tab1:
    uploaded_file = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
    if uploaded_file:
        try:
            image_input = Image.open(uploaded_file)
            st.image(image_input, caption="Uploaded Image", use_container_width=True)

            with st.spinner("Analyzing..."):
                result_img, confidence, label = detect_and_predict(image_input)

            st.image(result_img, caption="Detection Result", use_container_width=True)
            if confidence is not None:
                st.metric("Confidence", f"{confidence*100:.2f}%")
                if "Mask" in label:
                    st.success(label)
                else:
                    st.error(label)
            else:
                st.warning(label)

        except Exception as e:
            st.error(f"❌ Error: {str(e)}")

with tab2:
    camera_image = st.camera_input("Take a picture")
    if camera_image:
        try:
            image_input = Image.open(camera_image)
            st.image(image_input, caption="Webcam Snapshot", use_container_width=True)

            with st.spinner("Analyzing..."):
                result_img, confidence, label = detect_and_predict(image_input)

            st.image(result_img, caption="Detection Result", use_container_width=True)
            if confidence is not None:
                st.metric("Confidence", f"{confidence*100:.2f}%")
                if "Mask" in label:
                    st.success(label)
                else:
                    st.error(label)
            else:
                st.warning(label)

        except Exception as e:
            st.error(f"❌ Error: {str(e)}")