Preventing Structural Deterioration using Computer Vision

Recent research demonstrates how hyperspectral cameras and full-waveform laser scanners could soon be used to inspect the water content of concrete during construction. This is how to preventing structural deterioration using computer vision.

‍

It’s difficult to predict how long a concrete building will stave off the need for serious repairs. Thirty years? Sixty years? If the design is good and the materials are of high quality then perhaps the building will last. However, there is a crucial moment during the building’s construction that can make all the difference.

‍

It’s a hot day. Tempers are flaring. Due to coordination issues, several ready-mix trucks have been sitting on site for the better part of an hour. The concrete passes its tests and the pour finally begins. There is concern about meeting the discharge time requirement. The frustrations of viscous concrete make adding water irresistibly tempting. A laborer begins to argue the concrete is too dry. The fight escalates. Water-paranoid inspectors are hesitant to allow any water to be added on-site. Water-happy contractors consider adding water to be a God-given right.

‍

Nothing would be easier. We add a little excess mixing water, the workability of the concrete improves, and the concrete is placed on time. After all, nobody enjoys schlepping around viscous concrete or having to watch other people struggle. Unfortunately, things are not so simple. The uncontrolled addition of water is a serious gamble and causes structural deterioration. Whilst the short-term objective may be achieved, excess water is often the cause of many long-term undesirable effects in concrete:

‍

• Insufficient compressive strength
• Dusting, scaling, and streaking
• Cracking and permeability,
• Corrosion of reinforcing steel

‍

According to recently published research [1, 2], building stakeholders may soon have a new technique for quickly inspecting the integrity of in-place construction materials, e.g. identifying concrete with a higher-than-recommended water content prone to structual deterioriation. Specifically, the results demonstrate the feasibility of short-wave infrared spectrometry in assessing water-to-cement ratio and density of hardened concrete.

‍

“Resulting in an 89% correct classification for the 18 validation samples, thereby demonstrating that SWIR spectrometry can detect differences in initial water/cement ratios for hardened concretes.” [1]

‍

This capability is achieved using hyperspectral imaging. Hyperspectral images are produced by sensors called imaging spectrometers. Whereas a typical camera will record three bands of light per pixel: red, green, and blue; these spectrometers produce images where each individual pixel contains the light intensity for up to several hundred individual spectral bands. The resulting images can be used to characterize objects with unprecedented precision and minute detail. Imagine deploying these cameras on your construction site and having the assurance the concrete you’ve placed meets specifications and mitigate structural deterioration. At Avvir, we digitize mission-critical infrastructure and we’re excited about the potential of this new cutting-edge technology.

‍

[1] Zahiri, Z., Laefer, D. F., & Gowen, A. (2018). The feasibility of short-wave infrared spectrometry in assessing water-to-cement ratio and density of hardened concrete. Construction and Building Materials, 185, 661-669.

[2] Azadbakht, M., Fraser, C. S., & Khoshelham, K. (2016). Improved urban scene classification using full-waveform LiDAR. Photogrammetric Engineering & Remote Sensing, 82(12), 973-980.