Application of Computer Vision to Crack Detection of Concrete Structure

Su, Tung-Ching

doi:10.7763/ijet.2014.v5.596

Cited by 22 publications

(12 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some studies presented their methodologies based on machine learning and neural networks for identification and quantification of cracks starting from images collected by remote-controlled robots (Jahanshahi et al, 2013;Prasanna et al, 2016). Other similar studies based on images for the analysis of cracks have been presented (Lins and Givigi, 2016;Nguyen et al, 2014;Su, 2013;Valença et al, 2013). Usually computer vision functions are used to automatically extract the geometric properties of the damaged elements.…”

Section: Literary Reviewmentioning

confidence: 99%

Condition Assessment of Rc Bridges. Integrating Machine Learning, Photogrammetry and Bim

Borin

Cavazzini

2019

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

<p><strong>Abstract.</strong> The survey of building pathologies is focused on reading the state of conservation of the building, composed by the survey of constructive and decorative details, the masonry layering, the crack pattern, the degradation and the color recognition. The drawing of these representations is a time-consuming task, accomplished by manual work by skilled operators who often rely on in-situ analysis and on pictures. In this project three-dimensional an automated method for the condition survey of reinforced concrete spalling has been developed. To realize the automated image-based survey it has been exploited the Mask R-CNN neural network. The training phase has been executed over the original model, providing new examples of images with concrete cover detachments. At the same time, a photogrammetry process involved the images, in order to obtain a point cloud which acts as a reference to a Scan to BIM process. The BIM environment serves as a collector of information, as it owns the ontology to recreate entities and relationships. The information as extracted by neural network and photogrammetry serve to create the pictures which depict the concrete spalling in the BIM environment. A process of projecting information from the images to the BIM recreates the shapes of the pathology on the objects of the model, which becomes a decision support system for the built environment. A case study of a concrete beam bridge in northern Italy demonstrates the validity of the process.</p>

show abstract

Section: Literary Reviewmentioning

confidence: 99%

Condition Assessment of Rc Bridges. Integrating Machine Learning, Photogrammetry and Bim

Borin

Cavazzini

2019

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…In related studies on image processing techniques such as ones that applied the Otsu method [9], which is based on shooting variables and threshold value settings, to concrete structures such as roads, tunnels, and bridges to detect and measure cracks [10][11][12][13], image preprocessing studies have been conducted using histograms, mean values, median values, and Gaussian filters. Studies on feature extraction using morphology techniques [14][15][16] and studies on contour detection through primary and secondary differentiation [17,18] have also been conducted and repeatedly developed. Thereafter, the field of applicable target structures in studies conducted on crack detection was expanded to include bridges, tunnels, and dams that cannot be easily accessed.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Fine Crack Images Using Image Processing Technique and High-Resolution Camera

Jeong

Han

et al. 2021

Applied Sciences

View full text Add to dashboard Cite

Visual inspections are performed to investigate cracks in concrete infrastructure. These activities require manpower or equipment such as articulated ladders. Additionally, there are health and safety issues because some structures have low accessibility. To deal with these problems, crack measurement with digital images and digital image processing (DIP) techniques have been adopted in various studies. The objective of this experimental study is to evaluate the optical limit of digital camera lenses as working distance increases. Three different lenses and two digital cameras were used to capture images of lines ranging from 0.1 to 0.5 mm in thickness. As a result of the experiments, it was found that many elements affect width measurement. However, crack width measurement is dependent on the measured pixel values. To accurately measure width, the measured pixel values must be in decimal units, but that is theoretically impossible. According to the results, in the case of 0.3 mm wide or wider cracks, a working distance of 1 m was secured when the focal length was 50 mm, and working distances of 3 m and 4 m were secured when the focal length was 100 mm and 135 mm, respectively. However, for cracks not wider than 0.1 mm, focal lengths of 100 mm and 135 mm showed measurability within 1 m, but a focal length of 50 mm was judged to hardly enable measurement except for certain working positions. Field measurement tests were conducted to verify measurement parameters identified by the results of the indoor experiment. The widths of actual cracks were measured through visual inspection and used for the analysis. From the evaluation, it was confirmed that the number of pixels corresponding to the working distance had a great influence on crack width measurement accuracy when using image processing. Therefore, the optimal distance and measurement guidelines required for the measurement of the size of certain objects was presented for the imaging equipment and optical equipment applied in this study.

show abstract

“…Xu et al [70] resized the resolution of rectangular raw images of damaged reinforced concrete columns into smaller square pixels image to reduce calculation costs on the supervised learning procedure. Su [71] downsized 100 sampling images that were used in training model for concrete pavement to reduce computational time. Xu et al [72] resized the height pixel unit of grayscale images using bicubic interpolation then cropped them into smaller elements as the input of the deep network.…”

Section: Introductionmentioning

confidence: 99%

Implementation and Evaluation of Vision-Based Sensor Image Compression for Close-Range Photogrammetry and Structural Health Monitoring

Ngeljaratan

Moustafa

2020

Sensors

View full text Add to dashboard Cite

Much research is still underway to achieve long-term and real-time monitoring using data from vision-based sensors. A major challenge is handling and processing enormous amount of data and images for either image storage, data transfer, or image analysis. To help address this challenge, this study explores and proposes image compression techniques using non-adaptive linear interpolation and wavelet transform algorithms. The effect and implication of image compression are investigated in the close-range photogrammetry as well as in realistic structural health monitoring applications. For this purpose, images and results from three different laboratory experiments and three different structures are utilized. The first experiment uses optical targets attached to a sliding bar that is displaced by a standard one-inch steel block. The effect of image compression in the photogrammetry is discussed and the monitoring accuracy is assessed by comparing the one-inch value with the measurement from the optical targets. The second application is a continuous static test of a small-scale rigid structure, and the last application is from a seismic shake table test of a full-scale 3-story building tested at E-Defense in Japan. These tests aimed at assessing the static and dynamic response measurement accuracy of vision-based sensors when images are highly compressed. The results show successful and promising application of image compression for photogrammetry and structural health monitoring. The study also identifies best methods and algorithms where effective compression ratios up to 20 times, with respect to original data size, can be applied and still maintain displacement measurement accuracy.

show abstract

Application of Computer Vision to Crack Detection of Concrete Structure

Cited by 22 publications

References 17 publications

Condition Assessment of Rc Bridges. Integrating Machine Learning, Photogrammetry and Bim

Condition Assessment of Rc Bridges. Integrating Machine Learning, Photogrammetry and Bim

Analysis of Fine Crack Images Using Image Processing Technique and High-Resolution Camera

Implementation and Evaluation of Vision-Based Sensor Image Compression for Close-Range Photogrammetry and Structural Health Monitoring

Contact Info

Product

Resources

About