EYECON - Context-based Detection of Mixed-Criticality Events using Computer Vision


Fire:From lighter flames to accidental incidents, fires vary in severity from moderate to catastrophic, emphasizing the importance of readiness

Theft:Theft can range from minor incidents to major crimes, highlighting the need for vigilance and preventative measures

Crowd:From bustling crowds to packed events, crowd dynamics vary, emphasizing the importance of safety protocols and crowd management strategies

Traffic:Traffic accidents span from minor bumps to severe collisions, stressing the significance of safe driving habits.



Fire Incidents



  • Banks
  • Offices
  • industries
  • Petrol Pump
  • Shopping Plaza
  • Educational Institutes

How Does It Works?

Real-Time Data Capture:Our project seamlessly captures data in real-time.

Machine Learning Models: Cutting-edge ML models are employed to identify the context and criticality of events.

Context Identification:Events are categorized as moderate, critical, or catastrophic based on their context.

Versatile Application:Our system operates across various scenarios including fire incidents, traffic situations, crowd management, and theft detection..

Precision and Efficiency:By leveraging ML algorithms, our project ensures precise event identification and efficient response mechanisms.


Development Team

Alina Arshad Filza Akhlaq Syeda Ravia Ejaz Ayesha Zia Aiman Imran Muhammad Yahya

Faculty Team

Dr. Jawwad Shamsi Dr. Burhan Khan Dr. Narmeen Bawany