Recognition and identification of defects in civil structures

Whenever the same civil structure is examined by different examiners, all of them show different results. This is due to the factors like accessibility, light intensity, lack of specialised knowledge, perception of maintenance, visual acuity, and colour vision. Engineering structures are usu

Project Title

Recognition and identification of defects in civil structures

Project Area of Specialization

Artificial Intelligence

Project Summary

Whenever the same civil structure is examined by different examiners, all of them show different results. This is due to the factors like accessibility, light intensity, lack of specialised knowledge, perception of maintenance, visual acuity, and colour vision. Engineering structures are usually exposed to fatigue stress, continuous load which results in the process of degradation on the structure`s surface. Crack detection can be done with two procedures which are destructive testing non-Destructive testing using the modern tools of surface crack detection, surveying tools, and visual examination. With the advancement in technology on a daily basis, the increase in demand and efficiency, an image-based defect detection for structural inspection. Some of the irregularities such as random nature of the defect, illuminated conditions, shading blemishes and concrete spall in the acquired images may pose challenges in image-based detection.

Project Objectives

Defects are the earliest signs of degradation of any structure, the parameters such length, width, size and direction of propagation of defects impose harmful effects to the durability of the structure. Therefore, image-based defect detection is proposed as a fast and reliable alternative to manual inspection. It is safer and easier since it does not involve destructive means of inspection.

Project Implementation Method

As far as our model is concerned, a user provided image goes through seven filters and the output is displayed to the user. After reading the image, the algorithm converts the image into grayscale to observe severity of depth in three shades then, the image is blurred out to smoothen the edges of the crack then, a logarithmic filter is applied to identify areas of rapid change in image pixels. After that, a canny edge detector is used to highlight the edges. Lastly, the closing approach of the morphological filter is applied to sharpen the image. Following these conversions, an output image is then rendered to the landing page where users can estimate the parameters.

Benefits of the Project

image-based defect detection is proposed as a fast and reliable alternative to manual inspection. It is safer and easier since it does not involve destructive means of inspection.

Technical Details of Final Deliverable

Civil structural examination is commonly observed using destructive methods. In this method, the examiner examines the structure by the use of hammers or any other related tools. As a result, imposing more damage to the structure may even further reduce the durability of the structure.

Final Deliverable of the Project

Software System

Core Industry

IT

Other Industries

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com