Update app.py

app.py

@@ -1,25 +1,36 @@
-from PIL import Image
+import streamlit as st
 import numpy as np
 import cv2
+from PIL import Image
 import requests
 import face_recognition
+from keras.models import load_model
 import os
-import streamlit as st
 
 # Set page title and description
 st.set_page_config(
-    page_title="Aadhaar Based Face Recognition Attendance System",
+    page_title="Face Recognition Attendance System With Emotion Detection",
     page_icon="📷",
     layout="centered",
     initial_sidebar_state="collapsed"
 )
-st.title("Attendance System Using Face Recognition 📷")
-st.markdown("This app recognizes faces in an image, verifies Aadhaar card details, and updates attendance records with the current timestamp.")
+st.title("Attendance System Using Face Recognition and Emotion Detection 📷")
+st.markdown("This app recognizes faces in an image, detects emotions, and updates attendance records with the current timestamp.")
 
-# Load images for face recognition
-Images = []   # List to store Images
-classnames = []  # List to store classnames
-aadhar_numbers = []  # List to store Aadhaar numbers
+# Load emotion detection model
+@st.cache_resource
+def load_emotion_model():
+    model = load_model('CNN_Model_acc_75.h5')
+    return model
+
+emotion_model = load_emotion_model()
+
+# Emotion labels
+emotion_labels = ['angry', 'fear', 'happy', 'neutral', 'sad', 'surprise']
+
+# Load known faces and classnames
+Images = []   
+classnames = []  
 
 directory = "photos"
 myList = os.listdir(directory)
@@ -30,103 +41,81 @@ for cls in myList:
         curImg = cv2.imread(img_path)
         Images.append(curImg)
         classnames.append(os.path.splitext(cls)[0])
-        # Assume Aadhaar number is part of the image filename (e.g., "123456_john.jpg")
-        aadhar_numbers.append(cls.split('_')[0])
-
-# Function to validate Aadhaar card number
-def validate_aadhaar(aadhaar):
-    # Implement your Aadhaar card validation logic here
-    # For simplicity, let's assume any 6-digit number is a valid Aadhaar card
-    return len(aadhaar) == 6 and aadhaar.isdigit()
 
-# Function to update Aadhaar data
-def update_data(name, aadhaar_number):
+# Function to update attendance data
+def update_data(name, emotion):
     url = "https://huggingface.glitch.me"
     url1 = "/update"
-    data = {'name': name, 'aadhaar': aadhaar_number}
-    response = requests.post(url + url1, data=data)
     
-    if response.status_code == 200:
-        st.success("Data updated on: " + url)
-    else:
-        st.warning("Data not updated")
+    data = {'name': name, 'emotion': emotion}
+    try:
+        response = requests.post(url + url1, data=data)
+        if response.status_code == 200:
+            st.success("Attendance updated successfully!")
+        else:
+            st.warning("Failed to update attendance!")
+    except Exception as e:
+        st.error(f"Error updating attendance: {e}")
 
 # Function to display image with overlay
-def display_image_with_overlay(image, name):
-    # Add overlay to the image (e.g., bounding box and name)
-    # ...
-
-    # Apply styling with CSS
-    st.markdown('<style>img { animation: pulse 2s infinite; }</style>', unsafe_allow_html=True)
+def display_image_with_overlay(image, name, emotion):
+    cv2.putText(image, f"{name} is feeling {emotion}", (20, 30), cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 0, 0), 2)
     st.image(image, use_column_width=True, output_format="PNG")
 
-# Take input Aadhaar card details
-aadhaar_number = st.text_input("Enter your Last 6-digits Aadhaar Number:")
-    
-# Take picture using the camera 
-img_file_buffer = st.camera_input("Take a picture")
-
 # Load images for face recognition
 encodeListknown = [face_recognition.face_encodings(img)[0] for img in Images]
 
-if img_file_buffer is not None:
-    # Validate Aadhaar card number
-    if validate_aadhaar(aadhaar_number):
-        test_image = Image.open(img_file_buffer)
-        image = np.asarray(test_image)
-
-        imgS = cv2.resize(image, (0, 0), None, 0.25, 0.25)
-        imgS = cv2.cvtColor(imgS, cv2.COLOR_BGR2RGB)
-        facesCurFrame = face_recognition.face_locations(imgS)
-        encodesCurFrame = face_recognition.face_encodings(imgS, facesCurFrame)
-
-        name = "Unknown"  # Default name for unknown faces
-        match_found = False  # Flag to track if a match is found
-
-        # Checking if faces are detected
-        if len(encodesCurFrame) > 0:
-            for encodeFace, faceLoc in zip(encodesCurFrame, facesCurFrame):
-                # Assuming that encodeListknown is defined and populated in your code
-                matches = face_recognition.compare_faces(encodeListknown, encodeFace)
-                faceDis = face_recognition.face_distance(encodeListknown, encodeFace)
-                matchIndex = np.argmin(faceDis)
-
-                if matches[matchIndex]:
-                    name = classnames[matchIndex].upper()
-                    
-                    # Check if Aadhaar number is found in the database
-                    if aadhaar_number not in aadhar_numbers:
-                        st.error("Face recognized, but Aadhaar number not found in the database.")
-                    else:
-                        # Update data only if a known face is detected and Aadhaar number is valid
-                        update_data(name, aadhaar_number)
-                        match_found = True  # Set the flag to True
-                    
-                else:
-                    # Face recognized, but not matched with Aadhaar number
-                    st.error("Face recognized, but Aadhaar number does not match.")
-
-                y1, x2, y2, x1 = faceLoc
-                y1, x2, y2, x1 = y1 * 4, x2 * 4, y2 * 4, x1 * 4
-                cv2.rectangle(image, (x1, y1), (x2, y2), (0, 255, 0), 2)
-                cv2.rectangle(image, (x1, y2 - 35), (x2, y2), (0, 255, 0), cv2.FILLED)
-                cv2.putText(image, name, (x1 + 6, y2 - 6), cv2.FONT_HERSHEY_COMPLEX, 1, (255, 255, 255), 2)
-
-            display_image_with_overlay(image, name)
-
-            # Display the name corresponding to the entered Aadhaar number
-            if not match_found:
-                # Match Aadhaar number with the list
-                aadhar_index = aadhar_numbers.index(aadhaar_number) if aadhaar_number in aadhar_numbers else None
-                if aadhar_index is not None:
-                    st.success(f"Match found: {classnames[aadhar_index]}")
-                else:
-                    st.warning("Face not detected, and Aadhaar number not found in the database.")
-            else:
-                st.success(f"Face recognized: {name}")
+# Upload image using the file uploader
+img_file_buffer = st.file_uploader("Upload an image", type=["jpg", "jpeg", "png"])
 
+if img_file_buffer is not None:
+    test_image = Image.open(img_file_buffer)
+    image = np.asarray(test_image)
+
+    imgS = cv2.resize(image, (0, 0), None, 0.25, 0.25)
+    imgS = cv2.cvtColor(imgS, cv2.COLOR_BGR2RGB)
+    facesCurFrame = face_recognition.face_locations(imgS)
+    encodesCurFrame = face_recognition.face_encodings(imgS, facesCurFrame)
+
+    name = "Unknown"  # Default name for unknown faces
+    match_found = False  # Flag to track if a match is found
+
+    # Emotion detection part
+    face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
+    emotion = "Neutral"  # Default emotion
+
+    if len(encodesCurFrame) > 0:
+        for encodeFace, faceLoc in zip(encodesCurFrame, facesCurFrame):
+            # Emotion detection
+            y1, x2, y2, x1 = faceLoc
+            roi = imgS[y1:y2, x1:x2]
+            roi = cv2.resize(roi, (48, 48))  # Resize to fit model
+            roi = cv2.cvtColor(roi, cv2.COLOR_BGR2RGB)
+            roi = np.expand_dims(roi, axis=0) / 255.0  # Preprocess the image
+            emotion_predictions = emotion_model.predict(roi)
+            emotion = emotion_labels[np.argmax(emotion_predictions)]
+
+            # Face recognition logic
+            matches = face_recognition.compare_faces(encodeListknown, encodeFace)
+            faceDis = face_recognition.face_distance(encodeListknown, encodeFace)
+            matchIndex = np.argmin(faceDis)
+
+            if matches[matchIndex]:
+                name = classnames[matchIndex].upper()
+                update_data(name, emotion)
+                match_found = True
+                
+            y1, x2, y2, x1 = faceLoc
+            y1, x2, y2, x1 = y1 * 4, x2 * 4, y2 * 4, x1 * 4
+            cv2.rectangle(image, (x1, y1), (x2, y2), (0, 255, 0), 2)
+            cv2.rectangle(image, (x1, y2 - 35), (x2, y2), (0, 255, 0), cv2.FILLED)
+            cv2.putText(image, name, (x1 + 6, y2 - 6), cv2.FONT_HERSHEY_COMPLEX, 1, (255, 255, 255), 2)
+
+        display_image_with_overlay(image, name, emotion)
+
+        if match_found:
+            st.success(f"Face recognized: {name} and Emotion: {emotion}")
         else:
-            st.warning("No faces detected in the image. Face recognition failed.")
-
+            st.warning("Face not detected, or no match found in the database.")
     else:
-        st.error("Invalid Aadhaar card number. Please enter a valid 6-digit Aadhaar number.")
+        st.warning("No faces detected in the image. Face recognition failed.")