Advanced Imaging Techniques for Damage Characterization of Concrete

Concrete mix is one of the most widely used construction materials in the world. The mix includes crushed stones, gravel, and sand which are typically bound together with cement and water. The proportion of each component in concrete is determined based on the required properties for the construction work. The mix proportions are either nominal mix or design mix. The nominal mix of concrete is used for ordinary construction work such as a small residential structure. On the other hand, the design mix relies on the proportions finalized using laboratory tests which are performed to determine the compressive strength of the mixture. Additionally, there are various types of concrete like plain concrete, reinforced concrete, precast concrete, high- density concrete, ready mix concrete, decorative concrete and rapid set concrete. The varieties have grown into numbers due to the different requirements and applications, for example, ready-mix concrete is widely popular since it is a machine mix which has higher precision and large volume readily available at the construction site. Reinforced concrete is widely used for the construction of infrastructure projects like bridges, buildings, highways, dams, and power plants. Annually, billions of tons of concrete are used around the world for the construction of various types of structures.

Cracking in concrete structures is one of the most important and primary indicators of a structure's health. It indicates the deterioration in the strength of the structure and warns against possible failure. Generally, physical inspections are carried out to detect defects in structures for further rectification, repair and reinforcement as and when required. The physical inspection of cracks in concrete structures of various sizes can assist in determining the overall serviceability of the structure.

Image based automated or semi-automated detection of cracks has the potential to overcome the limitations of manual inspection. In alternative to physical inspection, images of critical locations of the structure can be captured by state-of-the-art image acquisition devices. The processing of images provides information about the current condition of structures. With recent advancements in computational algorithms, vision-based inspection is emerging as an efficient technique for monitoring the structural health of structures. Using the latest Artificial Intelligence (AI) based techniques, a better interpretation of collected data can be obtained automatically.

Computer Vision refers to a specific field that can process images and videos to extract meaningful information. According to IBM, computer vision is the field of AI that enables computers and systems to derive meaningful information from digital images, videos and other visual inputs and take actions or make recommendations based on that information. Both categories of applications cover aspects such as the identification of structural components, characterizing local and global visible damage, acquiring, storing and communicating structural images and detecting changes from a reference image.