• Intelligent Systems

Technologies to detect building facades defects

PI: Wai Choong Lawrence WONG, Yit Lin Michael CHEW, Ker-Wei Justin YEOH, Ernest Leslie Sidney ABBOTT, Marcelo Jr. Huibomhoa ANG, Kim Huat David CHUA, Wai Leong LEE, Janine Yin Ee WONG, Liang PAN, Man DING, Yuehan LIN, Er Peng Edwin LOH, Jia Min LIEW

Opportunity

Building facades are required to be inspected to ensure that the facades elements will not fall off and pose a danger to people. Human-based visual inspection is the most frequently adopted method practically to obtain defect information such as the existence, location, severity based on human's experience and expertise. However, most of the time, manual inspection is time-consuming, costly and often subjective as the results highly rely on different standards and the inspector's experience. Building facades inspection is a challenging task due to building height and complexity of facades elements, manual inspection processes are longer, more tedious and costly as it often requires the inspector to work at height using a scaffold, gondola or even rope access. Hence, there is an urgent need for a new and innovative method to reduce safety risks and improve overall efficiency.

The global facade market size was valued at USD 217.9 billion in 2020 and is expected to expand at a compound annual growth rate (CAGR) of 6.4% from 2021 to 2028.

Technology

3D Annotation Tool: A cutting-edge inspection application with a 3D annotation tool, enabling users to pinpoint image locations and identify defects on building facades. By overlaying annotations onto a 3D reconstructed model, precise defect localization becomes effortless.

AI and CV-Powered Workflow: This innovation employs Artificial Intelligence and Computer Vision algorithms to detect various building facade defects, such as cracks, spalling, delamination, efflorescence, and metal corrosion. It utilizes sensor fusion techniques, including digital and thermal cameras, along with a distance sensor for comprehensive defect evaluation.

Photogrammetry and Image Fusion: By merging data from multiple sources like Unmanned Aircraft, handheld cameras, and online 3D maps, this invention creates realistic 3D models of buildings. The inclusion of GNSS-acquired control points ensures high accuracy and efficiency in model generation.

Unmanned Aircraft (UA) Flight Planning: Leveraging UA technology for efficient building facade inspection, this method generates optimal flight paths in the world geodetic coordinate system. It maximizes coverage while considering no-fly zones, ensuring safe and reliable data acquisition for comprehensive defect inspection.

Document Status

Download

Technology Readiness Level (TRL)

6

Product prototype validated in intended environment

Applications & Advantages

  • 01

    An effective and efficient method for defect detection for efficient building facades inspection to detect facades defects.

  • 02

    Efficient and accurate identification of different types of building façade defects that aids façade inspection effectiveness and productivity.

  • 03

    Quick navigation of defect location and details in an interactive 3D experience.

  • 04

    Enables evolution of AI models to facilitate continuous improvement in detection accuracy.