Update app.py

app.py

@@ -1,148 +1,115 @@
 import streamlit as st
-import cv2
 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 time
-import cvlib as cv
-from cvlib.object_detection import draw_bbox
-
-# Set page config
-st.set_page_config(page_title="Face Mask Detection", layout="wide")
-
-# Initialize session state
-if "image_captured" not in st.session_state:
-    st.session_state.image_captured = None
-if "camera_key" not in st.session_state:
-    st.session_state.camera_key = str(time.time())
-
-# Header
-st.markdown("<h1 style='text-align: center;'>😷 Face Mask Detection</h1>", unsafe_allow_html=True)
-input_method = st.selectbox("Choose Input Method", ["Camera Capture", "Upload Image"])
-
-# Dummy classifier (replace this with actual ML model if needed)
-def classify_face_dummy(face_img):
-    mean = np.mean(face_img)
-    if mean % 2 < 1:
-        return "Mask", 0.90
-    else:
-        return "No Mask", 0.60
-
-def detect_and_classify_faces(img_np):
-    # Ensure color format is BGR
-    if img_np.shape[2] == 3:  # 3 channels
-        img_bgr = cv2.cvtColor(img_np, cv2.COLOR_RGB2BGR)
-    else:
-        img_bgr = img_np
-
-    faces, confidences = cv.detect_face(img_bgr)
-    results = []
-
-    height, width = img_np.shape[:2]
-
-    for i, face in enumerate(faces):
-        startX, startY = max(face[0], 0), max(face[1], 0)
-        endX, endY = min(face[2], width - 1), min(face[3], height - 1)
-
-        face_crop = img_bgr[startY:endY, startX:endX]
-
-        if face_crop.size == 0:
-            continue
-
-        label, conf = classify_face_dummy(face_crop)
-        results.append({
-            "box": [startX, startY, endX - startX, endY - startY],
-            "label": label,
-            "confidence": conf
-        })
-    return results
-
-# Layout
-col1, col2 = st.columns(2)
-
-# ------------------- CAMERA CAPTURE MODE -------------------
-if input_method == "Camera Capture":
-    with col1:
-        st.markdown("### 📸 Capturing Image")
-        camera_img = st.camera_input("Take a photo", key=st.session_state.camera_key)
-
-        if camera_img:
-            st.session_state.image_captured = camera_img
-
-            # Clear Button
-            if st.button("❌ Clear photo"):
-                st.session_state.image_captured = None
-                st.session_state.camera_key = str(time.time())
-                st.experimental_rerun()
-
-            image = Image.open(camera_img)
-            st.image(image, caption="Captured Image", use_container_width=True)
-
-    if st.session_state.image_captured:
-        with col2:
-            st.markdown("### 🧠 Detection Result")
-            image = Image.open(st.session_state.image_captured).convert("RGB")
-            img_np = np.array(image)
-
-            results = detect_and_classify_faces(img_np)
 
-            # Annotate image
-            for res in results:
-                x, y, w, h = res["box"]
-                label = res["label"]
-                conf = res["confidence"]
-                color = (0, 255, 0) if label == "Mask" else (0, 0, 255)
-
-                cv2.rectangle(img_np, (x, y), (x + w, y + h), color, 2)
-                cv2.putText(img_np, f"{label} ({conf*100:.2f}%)", (x, y - 10),
-                            cv2.FONT_HERSHEY_SIMPLEX, 0.6, color, 2)
+# Page config with improved UI layout
+st.set_page_config(
+    page_title="😷 Smart Face Mask Detection",
+    layout="wide",
+    page_icon="😷"
+)
+
+# Load the model with caching
+@st.cache_resource
+def load_model_cached():
+    return load_model("project_face_mask_detection.keras")
+
+model = load_model_cached()
+
+# Haar Cascade for face detection
+face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + "haarcascade_frontalface_default.xml")
+
+# Sidebar for app description and tips
+with st.sidebar:
+    st.title("🧠 About This App")
+    st.markdown("""
+    This app uses deep learning to detect whether a person is wearing a face mask.
+    
+    - Upload or capture an image.
+    - Get instant feedback.
+    - Built with Streamlit & Keras.
+    """)
+    st.info("Tip: Use well-lit images with clear faces for best results.")
+    st.markdown("---")
+    st.caption("📍 Developed by YourName • 2025")
+
+# Core detection function
+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"
+
+    # Drawing results
+    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, 0.6, color, 2)
+
+    return Image.fromarray(image_np), confidence, label
+
+# App Title
+st.markdown("<h1 style='text-align: center;'>😷 AI Face Mask Detection System</h1>", unsafe_allow_html=True)
+st.markdown("<p style='text-align: center;'>Upload or capture an image to analyze mask presence.</p>", unsafe_allow_html=True)
+
+# Dropdown for choosing input method
+input_choice = st.selectbox("Choose Input Method", ["📤 Upload Image", "📷 Use Webcam"])
+
+if input_choice == "📤 Upload Image":
+    uploaded_file = st.file_uploader("Choose an image file", type=["jpg", "jpeg", "png"])
+    if uploaded_file:
+        image_input = Image.open(uploaded_file)
+        st.image(image_input, caption="Uploaded Image", use_container_width=True)
+
+        with st.spinner("Analyzing with AI model..."):
+            result_img, confidence, label = detect_and_predict(image_input)
+
+        col1, col2 = st.columns(2)
+        with col1:
+            st.image(result_img, caption="Detection Output", use_container_width=True)
+        with col2:
+            if confidence is not None:
+                st.metric("Confidence Score", f"{confidence*100:.2f}%")
+                if "Mask" in label:
+                    st.success(label)
+                else:
+                    st.error(label)
+            else:
+                st.warning(label)
 
-            st.image(img_np, caption="Result Image", channels="RGB", use_container_width=True)
+elif input_choice == "📷 Use Webcam":
+    camera_image = st.camera_input("Take a picture using webcam")
+    if camera_image:
+        image_input = Image.open(camera_image)
+        st.image(image_input, caption="Webcam Snapshot", use_container_width=True)
 
-            for i, face in enumerate(results, 1):
-                label_color = "green" if face["label"] == "Mask" else "red"
-                st.markdown(
-                    f"**Face {i}:** <span style='color:{label_color}'>{face['label']}</span> ({face['confidence']*100:.2f}%)",
-                    unsafe_allow_html=True)
+        with st.spinner("Analyzing..."):
+            result_img, confidence, label = detect_and_predict(image_input)
 
-            if any(face["label"] == "No Mask" for face in results):
-                st.error("⚠️ One or more people not wearing a mask!")
-            else:
-                st.success("✅ All faces have masks.")
-
-# ------------------- UPLOAD IMAGE MODE -------------------
-elif input_method == "Upload Image":
-    with col1:
-        st.markdown("### 📁 Upload Image")
-        uploaded_img = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
-        if uploaded_img:
-            image = Image.open(uploaded_img).convert("RGB")
-            st.image(image, caption="Uploaded Image", use_container_width=True)
-
-    if uploaded_img:
+        col1, col2 = st.columns(2)
+        with col1:
+            st.image(result_img, caption="Detection Output", use_container_width=True)
         with col2:
-            st.markdown("### 🧠 Detection Result")
-            img_np = np.array(image)
-            results = detect_and_classify_faces(img_np)
-
-            for res in results:
-                x, y, w, h = res["box"]
-                label = res["label"]
-                conf = res["confidence"]
-                color = (0, 255, 0) if label == "Mask" else (0, 0, 255)
-
-                cv2.rectangle(img_np, (x, y), (x + w, y + h), color, 2)
-                cv2.putText(img_np, f"{label} ({conf*100:.2f}%)", (x, y - 10),
-                            cv2.FONT_HERSHEY_SIMPLEX, 0.6, color, 2)
-
-            st.image(img_np, caption="Result Image", channels="RGB", use_container_width=True)
-
-            for i, face in enumerate(results, 1):
-                label_color = "green" if face["label"] == "Mask" else "red"
-                st.markdown(
-                    f"**Face {i}:** <span style='color:{label_color}'>{face['label']}</span> ({face['confidence']*100:.2f}%)",
-                    unsafe_allow_html=True)
-
-            if any(face["label"] == "No Mask" for face in results):
-                st.error("⚠️ One or more people not wearing a mask!")
+            if confidence is not None:
+                st.metric("Confidence Score", f"{confidence*100:.2f}%")
+                if "Mask" in label:
+                    st.success(label)
+                else:
+                    st.error(label)
             else:
-                st.success("✅ All faces have masks.")
+                st.warning(label)