After configuring the streamlit-authenticator my streamlit-option-menu is not visible. When I click logout button, the menu is visible for a second or two and the app redirects to the login widget as usual. (Authenticator works fine.) Full code below.
import cv2
import os
from datetime import datetime
from keras_facenet import FaceNet
import numpy as np
import uuid
import pandas as pd
import csv
import pygame
from sklearn.metrics.pairwise import cosine_similarity
import streamlit as st
from streamlit_authenticator import Authenticate
from streamlit_option_menu import option_menu
import yaml
from yaml.loader import SafeLoader
from streamlit_authenticator.utilities.exceptions import LoginError
import modifyPerson
from activityHistory import display_entry_logs
from blacklist import show_blacklisted_persons
from personLogs import list_saved_persons
st.set_page_config(
page_title="EntryFace",
page_icon="🔐",
layout="wide"
)
logo_path = "assests/logo2.png"
title_path = "assests/title.png"
# Initialize Pygame mixer
pygame.mixer.init()
sound_played = False
warning_displayed = False
# Initialize columns
col1, col2 = st.columns(2)
# ------------------------- USER AUTHENTICATION ---------------------------------
with open('./assests/config.yaml') as file:
config = yaml.load(file, Loader=SafeLoader)
authenticator = Authenticate(
config['credentials'],
config['cookie']['name'],
config['cookie']['key'],
config['cookie']['expiry_days'],
)
# ------------- Creating a login widget -------------------------------------
try:
authenticator.login()
except LoginError as e:
st.error(e)
if st.session_state["authentication_status"]:
with st.sidebar:
st.image(logo_path, width=200)
st.image(title_path, use_column_width=True)
st.sidebar.write(f'Welcome *{st.session_state["name"]}*')
authenticator.logout()
# ------------- Navigation Bar -----------
selected = option_menu(
menu_title="Main Menu",
options=["Main Feed", "Update", "Blacklist Section", "Entry History", "Person Logs"],
icons=["camera-video", "pencil-square", "ban", "clock-history", "person-lines-fill"],
menu_icon="house",
default_index=0,
styles={
"container" : {"z-index" : "99999"}},
)
def play_sound():
pygame.mixer.music.load("./assests/alarm.wav")
pygame.mixer.music.play()
def show_notification(message, success=True):
notification_box = st.empty()
if success:
notification_box.success(f":white_check_mark: {message}")
else:
notification_box.error(message)
# Create a directory to store CSV files
CSV_DIR = "entry_logs"
os.makedirs(CSV_DIR, exist_ok=True)
def record_entry(name, entry_type, entry_time):
try:
# Create a CSV file with today's date as filename
today_date = datetime.now().strftime("%Y-%m-%d")
csv_filename = os.path.join(CSV_DIR, f"{today_date}.csv")
# Check if the CSV file already exists
file_exists = os.path.isfile(csv_filename)
# Check if the name already exists in the CSV file
name_exists = False
if file_exists:
df = pd.read_csv(csv_filename)
if name in df['Name'].values:
name_exists = True
# Check if the entry is late or on time for staff
status = ""
minutes_late = 0
if entry_type == "Staff":
if entry_time <= "09:00:00":
status = "On Time"
else:
status = "Late"
# Calculate minutes late
entry_time_obj = datetime.strptime(entry_time, "%H:%M:%S")
late_time_obj = datetime.strptime("09:00:00", "%H:%M:%S")
minutes_late = (entry_time_obj - late_time_obj).total_seconds() / 60
# Write entry to CSV file if name doesn't exist
if not name_exists:
with open(csv_filename, mode='a', newline='') as file:
writer = csv.writer(file)
# Write header if the file is newly created
if not file_exists:
writer.writerow(
['Name', 'Type', 'Entry Time', 'Exit Time', 'Duration (hours)', 'Status', 'Minutes Late'])
# Write entry for the current person
writer.writerow([name, entry_type, entry_time, '', '', status, minutes_late])
# Display message in the sidebar based on the entry type and status
if entry_type == "Staff":
if status == "Late":
st.sidebar.write(f"{name} is {int(minutes_late)} minutes late")
else:
st.sidebar.write(f"{name} is on time")
else:
st.sidebar.write("Entry time saved for:", name)
else:
print("Name already exists in the CSV file:", name)
except Exception as e:
print(f"Error occurred while creating the CSV file: {e}")
def update_exit_time(name, exit_time):
try:
# Create a CSV file with today's date as filename
today_date = datetime.now().strftime("%Y-%m-%d")
csv_filename = os.path.join(CSV_DIR, f"{today_date}.csv")
# Read the CSV file
df = pd.read_csv(csv_filename)
# Find the row corresponding to the person's name
row_index = df.index[df['Name'] == name].tolist()
# If the person's name is found in the CSV file
if row_index:
entry_time = datetime.strptime(df.at[row_index[0], 'Entry Time'], "%H:%M:%S")
# Check if the exit time is already recorded
if pd.isnull(df.at[row_index[0], 'Exit Time']):
# Calculate the time difference
time_difference = (datetime.now() - entry_time).seconds
# Check if the time difference is greater than 10 seconds
if time_difference > 10:
df.at[row_index[0], 'Exit Time'] = exit_time
df.at[row_index[0], 'Duration (hours)'] = calculate_duration(df.at[row_index[0], 'Entry Time'],
exit_time)
st.sidebar.write("Exit time updated successfully for:", name)
# Write updated entries back to CSV file
df.to_csv(csv_filename, index=False)
else:
print("Exit time already recorded for:", name)
else:
print("Name not found in CSV file:", name)
except Exception as e:
print(f"Error occurred while updating exit time: {e}")
def calculate_duration(entry_time, exit_time):
entry_datetime = datetime.strptime(entry_time, "%H:%M:%S")
exit_datetime = datetime.strptime(exit_time, "%H:%M:%S")
duration = exit_datetime - entry_datetime
duration_hours = duration.total_seconds() / 3600
return duration_hours
# Function to show a message when a blacklisted person is detected
def show_blacklist_message(name):
st.warning(f"Warning: {name} is blacklisted!")
st.toast(f"Warning: {name} is blacklisted!")
if selected == "Main Feed":
with col2:
st.title(" ")
st.title(" ")
# Input field for name
name_input = st.text_input("Enter the name:")
name_type = st.radio("Type:", ["Staff", "Visitor", "Special"])
print("Selected type:", name_type)
# Place the button in the main panel
capture_button = st.button("Capture 50 Faces")
with col1:
# Define the directory where embedding files are saved
SAVE_DIR = "captured_data"
# Define the size of the embeddings
embedding_size = 128 # For example, if the embeddings are 128-dimensional
st.title("Live Camera Feed")
FRAME_WINDOW = st.image([])
# Use Streamlit caching to initialize camera once
@st.cache(allow_output_mutation=True)
def initialize_camera():
return cv2.VideoCapture(0)
camera = initialize_camera()
# Directory to save captured data
SAVE_DIR = "captured_data"
os.makedirs(SAVE_DIR, exist_ok=True)
# Load saved embedding files
saved_embedding_files = []
for root, dirs, files in os.walk(SAVE_DIR):
for file in files:
if file.endswith("_embeddings.npy"):
saved_embedding_files.append(os.path.join(root, file))
# Initialize saved_embeddings as None
saved_embeddings = None
# Load saved embeddings if files are found
if saved_embedding_files:
saved_embeddings_list = []
for saved_embedding_file in saved_embedding_files:
embeddings = np.load(saved_embedding_file)
saved_embeddings_list.append(embeddings)
saved_embeddings = np.concatenate(saved_embeddings_list, axis=0)
# Check if saved_embeddings is None or empty
if saved_embeddings is None or saved_embeddings.size == 0:
print("No saved embeddings found.")
else:
print("Saved embeddings loaded successfully.")
frame_count = 0
MAX_FRAMES = 50
embedder = FaceNet()
face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
entry_recorded = False # Flag to check if entry is recorded
while True:
ret, frame = camera.read()
if not ret:
st.write('Camera disconnected!')
break
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
# Face detection using Haar Cascade
faces = face_cascade.detectMultiScale(gray, scaleFactor=1.3, minNeighbors=5)
for (x, y, w, h) in faces:
face_img = frame[y:y + h, x:x + w]
embedding = embedder.embeddings([face_img])[0]
# Compare face embeddings with saved embeddings
similarities = cosine_similarity([embedding], saved_embeddings)
# Inside the loop where we compare embeddings and display names
recognized = False
blacklist_status = False
max_similarity = 0.0
recognized_name = ""
for saved_embedding_file in saved_embedding_files:
saved_embeddings = np.load(saved_embedding_file)
similarities = cosine_similarity([embedding], saved_embeddings)
if np.max(similarities) > 0.7:
recognized = True
max_similarity = np.max(similarities)
embedding_filename = os.path.basename(saved_embedding_file)
recognized_name = embedding_filename.split("_")[0]
record_type = embedding_filename.split("_")[1]
# Get the folder name corresponding to the recognized person
folder_name = os.path.dirname(saved_embedding_file)
# Split the folder name by "_"
folder_parts = folder_name.split("_")
# Check if the third part of the folder name is "blacklisted"
if len(folder_parts) > 2 and folder_parts[3].lower() == "blacklisted":
blacklist_status = True
print(blacklist_status)
print(folder_parts[3])
print("Alarm")
cv2.rectangle(frame, (x, y), (x + w, y + h), (255, 0, 0),
3) # Red rectangle for recognized faces
cv2.putText(frame, f"{recognized_name} (Similarity: {max_similarity:.2f})", (x, y - 10),
cv2.FONT_HERSHEY_SIMPLEX, 0.6, (255, 0, 0), 2)
if not warning_displayed:
show_blacklist_message(recognized_name)
warning_displayed = True
if not sound_played:
play_sound()
sound_played = True
else:
sound_played = False
warning_displayed = False
break
# Update the entry_recorded variable when a new entry is recorded
if recognized and not blacklist_status is True:
name = recognized_name
entry_type = record_type
entry_time = datetime.now().strftime("%H:%M:%S")
record_entry(name, entry_type, entry_time)
# Calculate exit time
exit_time = datetime.now().strftime("%H:%M:%S")
# Update exit time if time difference is greater than 10 seconds
update_exit_time(name, exit_time)
# Check if the person is blacklisted and show message
cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 3) # Green rectangle for recognized faces
cv2.putText(frame, f"{recognized_name} (Similarity: {max_similarity:.2f})", (x, y - 10),
cv2.FONT_HERSHEY_SIMPLEX, 0.6, (0, 255, 0), 2)
if not recognized:
cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 0, 255),
2) # Blue rectangle for unrecognized faces
cv2.putText(frame, 'Unknown Person', (x, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 2)
FRAME_WINDOW.image(frame)
if capture_button and name_input:
name = name_input.strip().replace(" ",
"_") # Remove leading/trailing spaces and replace spaces with underscores
type = name_type.strip().replace(" ",
"_") # Remove leading/trailing spaces and replace spaces with underscores
# Generate a unique ID for this capture session
timestamp = datetime.now().strftime("%Y-%m-%d")
unique_id = uuid.uuid4().hex[:8]
# Include "notblacklisted" in the folder name
unique_folder_name = f"{name}_{type}_notblacklisted_{timestamp}_{unique_id}"
session_dir = os.path.join(SAVE_DIR, unique_folder_name)
os.makedirs(session_dir, exist_ok=True)
all_embeddings = []
while frame_count < MAX_FRAMES:
_, frame = camera.read()
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
# Face detection using Haar Cascade
faces = face_cascade.detectMultiScale(gray, scaleFactor=1.3, minNeighbors=5)
for (x, y, w, h) in faces:
face_img = frame[y:y + h, x:x + w]
embedding = embedder.embeddings([face_img])[0]
# Save the image of the face
cv2.imwrite(os.path.join(session_dir, f"face_{frame_count}.jpg"), face_img)
all_embeddings.append(embedding)
frame_count += 1
# Convert the list of embeddings to a numpy array
all_embeddings = np.array(all_embeddings)
# Save all embeddings to a single file inside the generated folder
embeddings_file_path = os.path.join(session_dir, f"{name}_{type}_notblacklisted_embeddings.npy")
np.save(embeddings_file_path, all_embeddings)
# Show notification in the sidebar
show_notification(f"{name} ({type}) has been added")
capture_button = False
if selected == "Update":
# Define the directory where files are saved
SAVE_DIR = "captured_data"
# List all files in the directory
all_files = os.listdir(SAVE_DIR)
# Select a file to edit or delete
selected_file = st.selectbox("Select File to Edit/Delete", all_files)
# Input fields to enter new name and type
new_name = st.text_input("Enter New Name", value=selected_file.split("_")[0])
new_type = st.selectbox("Select New Type", options=["Staff", "Visitor", "Special"],
index=["Staff", "Visitor", "Special"].index(selected_file.split("_")[1]))
# Checkbox to select whether the entry is blacklisted or not
blacklisted = st.checkbox("Blacklisted")
# Button to save changes
if st.button("Save Changes"):
# Call the function to edit and save file details
modifyPerson.edit_entry_details(SAVE_DIR, selected_file, new_name, new_type, blacklisted)
# Checkbox to select whether to delete the selected person and folder
delete_checkbox = st.checkbox("Delete Person and Folder")
# Button to confirm deletion
if delete_checkbox:
if st.button("Confirm Deletion"):
# Call the function to delete the selected person and folder
modifyPerson.delete_person_and_folder(SAVE_DIR, selected_file)
if selected == "Blacklist Section":
st.title("Blacklist Section")
SAVE_DIR = "captured_data"
ENTRY_LOGS_DIR = "entry_logs"
show_blacklisted_persons(SAVE_DIR, ENTRY_LOGS_DIR)
if selected == "Entry History":
st.title("Entry Logs Viewer")
# Date input widget to select the date
selected_date = st.date_input("Select Date", datetime.now())
# Call the function to display entry logs for the selected date
display_entry_logs(selected_date)
if selected == "Person Logs":
st.title("List of Saved Persons")
# Define the directories where files and entry logs are saved
SAVE_DIR = "captured_data"
ENTRY_LOGS_DIR = "entry_logs"
# Input field to filter by type
filter_type = st.selectbox("Filter by Type", options=["All", "Staff", "Visitor", "Special"])
# List all saved persons in a table
if filter_type == "All":
st.write(list_saved_persons(SAVE_DIR, ENTRY_LOGS_DIR))
else:
st.write(list_saved_persons(SAVE_DIR, ENTRY_LOGS_DIR, filter_type))
elif st.session_state["authentication_status"] is False:
st.error('Username/password is incorrect')
elif st.session_state["authentication_status"] is None:
st.warning('Please enter your username and password')
# Saving config file
with open('../config.yaml', 'w', encoding='utf-8') as file:
yaml.dump(config, file, default_flow_style=False)
