Smart Street Surveillance: Real-Time People Flow Counter

Project Description:

Smart Street Surveillance is an AI-powered computer vision system that detects, tracks, and counts individuals in real-time across a public street or facility entrance using pre-trained YOLO object detection and tracking algorithms. The project is designed to distinguish when a person enters or exits a monitored zone by analyzing their trajectory with respect to a virtual boundary line (entry/exit line). It uses deep learning and live video feeds to annotate, log, and visualize human movement for smart urban management, public safety, and facility flow optimization.

Core Objectives:

• Detect humans in outdoor or open environments in real time.

• Track individual movement across video frames using unique IDs.

• Determine if a person has entered or exited a region based on a virtual line.

• Visually annotate the video feed with bounding boxes, tracking markers, and entry/exit counts.

• Log events with timestamps and track IDs for auditing and analytics.

Key Features:

✅ YOLO Object Detection: Detects people with high accuracy in diverse lighting and crowded environments.

✅ Line-Crossing Logic: Determines entry/exit events based on movement across a predefined virtual line.

✅ Real-Time Tracking: Uses persistent tracking to maintain identity across frames.

✅ Entry/Exit Counter: Live overlay showing total count of entered and exited individuals.

✅ Event Logging: Automatically logs every entry/exit event with timestamp and person ID.

✅ Customizable Line Placement: Adjust virtual line to suit any camera angle or scene.

✅ Optimized for CCTV Footage: Works well with low-angle street surveillance video.

Technologies Used:

• Ultralytics YOLO: For object detection and ID-based tracking.

• OpenCV: For frame processing, video handling, line drawing, and annotation.

• Python: Core language for logic, processing, and integration.

• Pandas: For structured logging and CSV file generation.

• NumPy: For efficient pixel and geometric calculations.

Use Cases:

• 🚧 Smart Cities: Monitor pedestrian traffic in streets, intersections, and sidewalks.

• 🏢 Corporate Campuses: Count people entering and exiting office buildings.

• 🏫 Schools & Colleges: Track student flow through gates.

• 🏥 Hospitals: Monitor patient/visitor movement in high-risk areas.

• 🏗️ Construction Sites: Ensure safety compliance by monitoring unauthorized entry.

• 🏘️ Gated Communities: Replace manual gate logs with automated digital tracking.

Customize Enhancements:

🚀 Face Recognition Add-on: Identify individuals in addition to counting.

📦 Database Integration: Sync logs with cloud or on-premise databases for long-term storage.

📈 Analytics Dashboard: Visualize hourly/daily trends, peak hours, and heatmaps.

🔔 Alert System: Trigger real-time alerts if suspicious activity is detected.

🧠 Activity Recognition: Extend model to detect behaviors (e.g., loitering, running).

📱 Mobile App or Web Panel: Remote access to live video and analytics.

Conclusion:

Smart Street Surveillance: Real-Time People Flow Counter brings powerful AI-driven pedestrian monitoring to urban and commercial spaces. By leveraging the YOLO model’s speed and accuracy, it provides a real-time, low-cost alternative to traditional manual monitoring systems. With its scalable design and rich logging features, this project can be deployed in smart city infrastructure, event monitoring, or enterprise security systems to help manage crowd movement efficiently and safely.