A Stand-Alone Smart Camera System for Online Post-Earthquake Building Safety Assessment.

Area of Science:

• Structural Engineering
• Computer Vision
• Seismic Monitoring

Background:

• Traditional computer vision for structural health monitoring faces computational challenges with numerous cameras.
• Manual processing of video data for post-earthquake safety assessment is time-consuming and impractical for large-scale events.
• A need exists for decentralized, automated systems for prompt building safety evaluation.

Purpose of the Study:

• To develop a stand-alone smart camera system for accurate interstory drift measurement.
• To enable decentralized and fully automatic building safety assessment immediately after seismic events.
• To overcome the computational and logistical limitations of centralized monitoring systems.

Main Methods:

• Developed a smart camera system integrating a camera, single-board computer, accelerometers, and microcontroller.
• Implemented algorithms for rotational effect compensation, target tracking, and sensor data fusion (camera and accelerometer).
• Embedded artificial neural networks and signal processing techniques for online data analysis and interstory drift calculation.

Main Results:

• The system achieved online processing of video and acceleration data using decentralized resources.
• Accurate measurement of interstory drifts, even with residual displacements, was demonstrated through sensor fusion.
• Feasibility was validated via large-scale shaking table tests on a steel building structure.

Conclusions:

• The proposed smart camera system offers a promising solution for rapid, automated post-earthquake building safety assessment.
• Decentralized processing significantly reduces computational burden compared to centralized approaches.
• The system's ability to provide immediate safety evaluations enhances post-disaster decision-making and response.