Inspecting Buildings Using Drones and Computer Vision: A Machine Learning Approach to Detect Cracks and Damages

Enhancing surface drainage mapping in eastern Canada with deep learning applied to LiDAR-derived elevation data

Agricultural dykelands in Nova Scotia rely heavily on a surface drainage technique called land forming, which is used to alter the topography of fields to improve drainage. The presence of land-formed fields provides useful information to better understand land utilization on these lands vulnerable to rising sea levels. Current field boundaries delineation and classification methods, such as manual digitalization and traditional segmentation techniques, are labour-intensive and often require manual and time-consuming parameter selection. In recent years, deep learning (DL) techniques, including convolutional neural networks and Mask R-CNN, have shown promising results in object recognition, image classification, and segmentation tasks. However, there is a gap in applying these techniques to detecting surface drainage patterns on agricultural fields. This paper develops and tests a Mask R-CNN model for detecting land-formed fields on agricultural dykelands using LiDAR-derived elevation data. Specifically, our approach focuses on identifying groups of pixels as cohesive objects within the imagery, a method that represents a significant advancement over pixel-by-pixel classification techniques. The DL model developed in this study demonstrated a strong overall performance, with a mean Average Precision (mAP) of 0.89 across Intersection over Union (IoU) thresholds from 0.5 to 0.95, indicating its effectiveness in detecting land-formed fields. Results also revealed that 53% of Nova Scotia's dykelands are being used for agricultural purposes and approximately 75% (6924 hectares) of these fields were land-formed. By applying deep learning techniques to LiDAR-derived elevation data, this study offers novel insights into surface drainage mapping, enhancing the capability for precise and efficient agricultural land management in regions vulnerable to environmental changes.

Assessment of Convolutional Neural Network Pre-Trained Models for Detection and Orientation of Cracks

Failure due to cracks is a major structural safety issue for engineering constructions. Human examination is the most common method for detecting crack failure, although it is subjective and time-consuming. Inspection of civil engineering structures must include crack detection and categorization as a key component of the process. Images can automatically be classified using convolutional neural networks (CNNs), a subtype of deep learning (DL). For image categorization, a variety of pre-trained CNN architectures are available. This study assesses seven pre-trained neural networks, including GoogLeNet, MobileNet-V2, Inception-V3, ResNet18, ResNet50, ResNet101, and ShuffleNet, for crack detection and categorization. Images are classified as diagonal crack (DC), horizontal crack (HC), uncracked (UC), and vertical crack (VC). Each architecture is trained with 32,000 images equally divided among each class. A total of 100 images from each category are used to test the trained models, and the results are compared. Inception-V3 outperforms all the other models with accuracies of 96%, 94%, 92%, and 96% for DC, HC, UC, and VC classifications, respectively. ResNet101 has the longest training time at 171 min, while ResNet18 has the lowest at 32 min. This research allows the best CNN architecture for automatic detection and orientation of cracks to be selected, based on the accuracy and time taken for the training of the model.

Segmentation of trabecular bone microdamage in Xray microCT images using a two-step deep learning method

INTRODUCTION

One of the current approaches to improve our understanding of osteoporosis is to study the development of bone microdamage under mechanical loading. The current practice for evaluating bone microdamage is to quantify damage volume from images of bone samples stained with a contrast agent, often composed of toxic heavy metals and requiring long tissue preparation. This work aims to evaluate the potential of linear microcracks detection and segmentation in trabecular bone samples using well-known deep learning models, namely YOLOv4 and Unet, applied on microCT images.

METHODS

Six trabecular bovine bone cylinders underwent compression until ultimate stress and were subsequently imaged with a microCT at a resolution of 1.95 μm. Two of these samples (samples 1 and 2) were then stained using barium sulfate (BaSO₄) and imaged again. The unstained samples (samples 3-6) were used to train two neural networks YOLOv4 to detect regions with microdamage further combined with Unet to segment the microdamage at the pixel level in the detected regions. Four different model versions of YOLOv4 were compared using the average Intersection over Union (IoU) and the mean average precision (mAP). The performance of Unet was also measured using two segmentation metrics, the Dice Score and the Intersection over Union (IoU). A qualitative comparison was finally done between the deep learning and the contrast agent approaches.

RESULTS

Among the four versions of YOLOv4, the YOLOv4p5 model resulted in the best performance with an average IoU of 45,32% and 51,12% and a mAP of 28.79% and 46.22%, respectively for samples 1 and 2. The segmentation performance of Unet provided better IoU and DICE score on sample 2 compared to sample 1. The poorer performance of the test on sample 1 could be explained by its poorer contrast to noise ratio (CNR). Indeed, sample 1 resulted in a CNR of 7,96, which was worse than the average CNR in the training samples, while sample 2 resulted in a CNR of 10,08. The qualitative comparison between the contrast agent and the deep learning segmentation showed that two different regions were segmented by the two techniques. Deep learning is segmenting the region inside the cracks while the contrast agent segments the region around it or even regions with no visible damage.

CONCLUSION

The combination of YOLOv4 for microdamage detection with Unet for damage segmentation showed a potential for the detection and segmentation of microdamage in trabecular bone. The accuracy of both neural networks achieved in this work is acceptable considering it is their first application in this specific field and the amount of data was limited. Even if the errors from both neural networks are accumulated, the two-steps approach is faster than the semantic segmentation of the whole volume.

Orthofaçade-Based Assisted Inspection Method for Buildings

Building façade assessment could be performed in a more efficient way using a multidisciplinary approach and modern technologies. This study proposes the orthofaçade-based assisted inspection method (AIM), universal and applicable to different types of façade cladding and suitable for application in the condition assessment of inaccessible building façades or high-rise and large structures of all kinds. The AIM method offers a multidisciplinary approach by combining unmanned aerial vehicle (UAV) technology, electronic tachymetry, and digital image processing techniques (photogrammetry and open-source computer vision methods). The method was verified in a case study performed on a high-rise building façade. On-site data acquisition of high-resolution images of façade and control points was conducted by UAV and tachymetry. The data were further processed in photogrammetric software in order to generate a georeferenced orthofaçade. Crack detection was performed at pixel level via computer code using the OpenCV library methods. The established diagnostic model, defined by control points, enables precise determination of crack location. Crack length, width, or area could be calculated based on the coordinates of its points, by performing simple mathematical operations. The AIM method provides automation of crack detection and precise determination of location and geometrical parameters of detected crack.

Big Data in Construction: Current Applications and Future Opportunities

Big data have become an integral part of various research fields due to the rapid advancements in the digital technologies available for dealing with data. The construction industry is no exception and has seen a spike in the data being generated due to the introduction of various digital disruptive technologies. However, despite the availability of data and the introduction of such technologies, the construction industry is lagging in harnessing big data. This paper critically explores literature published since 2010 to identify the data trends and how the construction industry can benefit from big data. The presence of tools such as computer-aided drawing (CAD) and building information modelling (BIM) provide a great opportunity for researchers in the construction industry to further improve how infrastructure can be developed, monitored, or improved in the future. The gaps in the existing research data have been explored and a detailed analysis was carried out to identify the different ways in which big data analysis and storage work in relevance to the construction industry. Big data engineering (BDE) and statistics are among the most crucial steps for integrating big data technology in construction. The results of this study suggest that while the existing research studies have set the stage for improving big data research, the integration of the associated digital technologies into the construction industry is not very clear. Among the future opportunities, big data research into construction safety, site management, heritage conservation, and project waste minimization and quality improvements are key areas.

An AI/ML-Based Strategy for Disaster Response and Evacuation of Victims in Aged Care Facilities in the Hawkesbury-Nepean Valley: A Perspective

The Hawkesbury-Nepean Valley, Australia’s longest coastal catchment, is spanned by a river system of more than 470 km, that runs from Goulburn to Broken Bay, covering a total area of over 2.2 million hectares. This region has remained prone to flood events, with considerable mortalities, economic impacts and infrastructural losses occurring quite regularly. The topography, naturally variable climatic conditions and the ‘bathtub’ effect in the region are responsible for the frequent flood events. In response, the Government at the national/federal, state and local level has focused on the design of efficient flood risk management strategies with appropriate evacuation plans for vulnerable communities from hospitals, schools, childcare and aged care facilities during a flood event. Despite these overarching plans, specialized response and evacuation plans for aged care facilities are critical to reducing the loss incurred by flood events in the region. This is the focus of this present paper, which reviews the history of flood events and responses to them, before examining the utilization of artificial intelligence (AI) techniques during flood events to overcome the flood risks. An early flood warning system, based on AI/Machine Learning (ML) strategy is being suggested for a timely decision, enhanced disaster prediction, assessment and response necessary to overcome the flood risks associated with aged care facilities within the Hawkesbury-Nepean region. A framework entailing AI/ML methods for identifying the safest route to the destination using UAV and path planning has been proposed for timely disaster response and evacuation of the residents of aged care facilities.