Search Results (16)

This study investigates the engineering characteristics of a newly commissioned higher education building through the Bioharmological Conformity Assessment (BCA) method, specifically using the 2020vEB version. The BCA is a novel evaluation approach that assesses whether a building aligns with the identity of its users and its intended function. The engineering attributes of the structure were assessed across 12 core criteria, encompassing a total of 600 individual parameters. Findings from the BCA inspection indicate that the newly completed building falls into the category of “Near-Standard Building/Minor Modifications Required.” The BCA score was calculated as 398.73, corresponding to a deficiency rate of 25.50%. Notably, significant shortcomings were observed in categories such as user identity and intended use, Physical Characteristics of the Space, and Ecological and Seismological Suitability. Consequently, targeted improvements are necessary to align the building with bioharmological principles, requiring only minor adjustments to rectify the identified deficiencies. Full article

Structural adhesives or fixing glues are typically applied to larger components on printed circuit boards (PCBs) to increase mechanical stability and minimize damage from vibration. Existing work tends to focus on component placement verification and solder joint analysis, etc. However, the detection of structural adhesives remains largely unexplored. This paper proposes a vision-based method for detecting structural adhesive defects on PCBs. The method uses HSV color segmentation to extract PCB regions, followed by Hough-transform-based morphological analysis to identify board features. The perspective transformation then extracts and rectifies the adhesive regions, and constructs an adhesive region template by detecting the standard adhesive area ratio in its corresponding adhesive region. Finally, template matching is used to detect the structural adhesives. The experimental results show that this approach can accurately detect the adhesive state of PCBs and identify the qualified/unqualified locations, providing an effective vision-based detection scheme for PCB manufacturing. The main contributions of this paper are as follows: (1) A vision-based structural adhesive detection method is proposed, and its detailed algorithm is presented. (2) The developed system includes a user-friendly visualization interface, streamlining the inspection workflow. (3) Actual experiments are performed to evaluate this study, and the results validate its effectiveness. Full article

The persistence of recurrent safety noncompliance (RSN) in university laboratories presents a critical challenge to laboratory safety risk management. This paper deconstructs RSN by conducting an in-depth analysis of potential safety risks, their underlying causes, and management obstacles. The research reveals that the phenomenon of RSN is fundamentally the result of the combined effects of complex human factor risks and outdated management methods. At the human factor level, cognitive biases regarding experimental safety risks and negative resistance lead to “habitual violations” of safety regulations. At the management level, routine laboratory safety inspections, requirements for rectifying safety hazards, and commonly adopted punitive measures have proven insufficient to prevent RSN. To address this issue, this study proposes actively leveraging the advantages of artificial intelligence (AI) in dynamic perception and proactive interventions. It advocates for the deep integration of AI technologies into the transformation of the management paradigm for RSN in university laboratories. Furthermore, this study preliminarily explores the application prospects, applicable principles, and scope of application of AI technologies in this context, providing an important reference for enhancing the systematic management of RSN in university laboratories. Full article

This paper presents an innovative approach that utilizes infused images from vibration signals and visual inspections to enhance the efficiency and accuracy of structure health monitoring through GoogLeNet. Scrutiny of the structure of GoogLeNet identified four key parameters, and thus, the optimization of GoogLeNet was completed through manipulation of the four key parameters. First, the impact of the number of inception modules on the performance of GoogLeNet revealed that employing eight inception layers achieves remarkable 100% accuracy while requiring less computational time compared to nine layers. Second, the choice of activation function was studied, with the Rectified Linear Unit (ReLU) emerging as the most effective option. Types of optimizers were then researched, which identified Stochastic Gradient Descent with Momentum (SGDM) as the most efficient optimizer. Finally, the influence of learning rate was compared, which found that a rate of 0.001 produces the best outcomes. By amalgamating these findings, a comprehensive optimized GoogLeNet model was found to identify damage cases effectively and accurately through infused images from vibrations and visual inspections. Full article

Vehicle exterior inspection is a critical operation for identifying defects and ensuring the overall safety and integrity of vehicles. Visual-based inspection of moving objects, such as vehicles within dynamic environments abounding with reflections, presents significant challenges, especially when time and accuracy are of paramount importance. Conventional exterior inspections of vehicles require substantial labor, which is both costly and prone to errors. Recent advancements in deep learning have reduced labor work by enabling the use of segmentation algorithms for defect detection and description based on simple RGB camera acquisitions. Nonetheless, these processes struggle with issues of image orientation leading to difficulties in accurately differentiating between detected defects. This results in numerous false positives and additional labor effort. Estimating image poses enables precise localization of vehicle damages within a unified 3D reference system, following initial detections in the 2D imagery. A primary challenge in this field is the extraction of distinctive features and the establishment of accurate correspondences between them, a task that typical image matching techniques struggle to address for highly reflective moving objects. In this study, we introduce an innovative end-to-end pipeline tailored for efficient image matching and stitching, specifically addressing the challenges posed by moving objects in static uncalibrated camera setups. Extracting features from moving objects with strong reflections presents significant difficulties, beyond the capabilities of current image matching algorithms. To tackle this, we introduce a novel filtering scheme that can be applied to every image matching process, provided that the input features are sufficient. A critical aspect of this module involves the exclusion of points located in the background, effectively distinguishing them from points that pertain to the vehicle itself. This is essential for accurate feature extraction and subsequent analysis. Finally, we generate a high-quality image mosaic by employing a series of sequential stereo-rectified pairs. Full article

The success of regional bridge condition assessment, a crucial component of systematic maintenance strategies, relies heavily on comprehensive, well-structured regional bridge databases. This study proposes the data extraction, cleaning, and integration method for the construction of such databases. First, this research proposes an extraction method tailored for unstructured data often present in inspection reports. Additionally, this paper meticulously outlines a cleaning procedure designed to rectify two distinct categories of typical errors that are present within the inspection data. Subsequently, this study takes a holistic approach by establishing integration rules that harmonize data from various sources, including inspection records, monitoring data, traffic statistics, as well as design and construction blueprints. The architectural framework of the regional bridge information database is then meticulously laid out. To validate and demonstrate the effectiveness of the method, this study applies them to a set of representative highway bridges situated within Shandong Province. The results show that this approach can be used to successfully establish a functional regional bridge database. The database plays a pivotal role in harnessing the latent potential of an extensive range of multi-source information and propels the field of bridge condition assessment forward by providing a solid basis for informed decision making and strategic planning in the realm of infrastructure maintenance. Full article

When a mobile robot inspects tasks with complex requirements indoors, the traditional backstepping method cannot guarantee the accuracy of the trajectory, leading to problems such as the instrument not being inside the image and focus failure when the robot grabs the image with high zoom. In order to solve this problem, this paper proposes an adaptive backstepping method based on double Q-learning for tracking and controlling the trajectory of mobile robots. We design the incremental model-free algorithm of Double-Q learning, which can quickly learn to rectify the trajectory tracking controller gain online. For the controller gain rectification problem in non-uniform state space exploration, we propose an incremental active learning exploration algorithm that incorporates memory playback as well as experience playback mechanisms to achieve online fast learning and controller gain rectification for agents. To verify the feasibility of the algorithm, we perform algorithm verification on different types of trajectories in Gazebo and physical platforms. The results show that the adaptive trajectory tracking control algorithm can be used to rectify the mobile robot trajectory tracking controller’s gain. Compared with the Backstepping-Fractional-Older PID controller and Fuzzy-Backstepping controller, Double Q-backstepping has better robustness, generalization, real-time, and stronger anti-disturbance capability. Full article

The deterioration of infrastructure’s health has become more predominant on a global scale during the 21st century. Aging infrastructure as well as those structures damaged by natural disasters have prompted the research community to improve state-of-the-art methodologies for conducting Structural Health Monitoring (SHM). The necessity for efficient SHM arises from the hazards damaged infrastructure imposes, often resulting in structural collapse, leading to economic loss and human fatalities. Furthermore, day-to-day operations in these affected areas are limited until an inspection is performed to assess the level of damage experienced by the structure and the required rehabilitation determined. However, human-based inspections are often labor-intensive, inefficient, subjective, and restricted to accessible site locations, which ultimately negatively impact our ability to collect large amounts of data from inspection sites. Though Deep-Learning (DL) methods have been heavily explored in the past decade to rectify the limitations of traditional methods and automate structural inspection, data scarcity continues to remain prevalent within the field of SHM. The absence of sufficiently large, balanced, and generalized databases to train DL-based models often results in inaccurate and biased damage predictions. Recently, Generative Adversarial Networks (GANs) have received attention from the SHM community as a data augmentation tool by which a training dataset can be expanded to improve the damage classification. However, there are no existing studies within the SHM field which investigate the performance of DL-based multiclass damage identification using synthetic data generated from GANs. Therefore, this paper investigates the performance of a convolutional neural network architecture using synthetic images generated from a GAN for multiclass damage detection of concrete surfaces. Through this study, it was determined the average classification performance of the proposed CNN on hybrid datasets decreased by 10.6% and 7.4% for validation and testing datasets when compared to the same model trained entirely on real samples. Moreover, each model’s performance decreased on average by 1.6% when comparing a singular model trained with real samples and the same model trained with both real and synthetic samples for a given training configuration. The correlation between classification accuracy and the amount and diversity of synthetic data used for data augmentation is quantified and the effect of using limited data to train existing GAN architectures is investigated. It was observed that the diversity of the samples decreases and correlation increases with the increase in the number of synthetic samples. Full article

This paper presents an automated inspection and classification system for automotive component remanufacturing industry, based on ensemble learning. The system is based on different stages allowing to classify the components as good, rectifiable or rejection according to the manufacturer criteria. A study of two deep learning-based models’ performance when used individually and when using an ensemble of them is carried out, obtaining an improvement of 7% in accuracy in the ensemble. The results of the test set demonstrate the successful performance of the system in terms of component classification. Full article

Sanitary inspections (SIs) are checklists of questions used for achieving/maintaining the safety of drinking-water supplies by identifying observable actual and potential sources and pathways of contamination. Despite the widespread use of SIs, the effects of training on SI response are understudied. Thirty-six spring supplies were inspected on two occasions, pre- and post-training, by an instructor from the research team and four local inspectors in the Mukono District of Uganda. SI score agreement between the instructor and each inspector was calculated using Lin’s concordance correlation coefficient. Average SI score agreement between the instructor and all inspectors increased post-training for the Yes/No answer type (0.262 to 0.490). For the risk level answer type (e.g., No, Low, Medium, High), average SI score agreement between the instructor and all inspectors increased post-training (0.301 to 0.380). Variability of SI scores between the four inspectors was calculated using coefficient of variation analysis. Average SI score variability between inspectors reduced post-training for both answer types, Yes/No (21.25 to 16.16) and risk level (24.12 to 19.62). Consistency of answer agreement between the four inspectors for each individual SI question was calculated using index of dispersion analysis. Average answer dispersion between inspectors reduced post-training for both answer types, Yes/No (0.41 to 0.27) and risk level (0.55 to 0.41). The findings indicate that training has a positive effect on improving answer agreement between inspectors. However, advanced training or tailoring of SI questions to the local context may be required where inconsistency of responses between inspectors persists, especially for the risk level answer type that requires increased use of inspector risk perception. Organisations should be aware of the potential inconsistency of results between inspectors so that this may be rectified with appropriate training and, where necessary, better SI design and customisation. Full article

This paper mainly improves the visual geometry group network-16 (VGG-16), which is a classic convolutional neural network (CNN), to classify the surface defects on cement concrete bridges in an accurate manner. Specifically, the number of fully connected layers was reduced by one, and the Softmax classifier was replaced with a Softmax classification layer with seven defect tags. The weight parameters of convolutional and pooling layers were shared in the pre-trained model, and the rectified linear unit (ReLU) function was taken as the activation function. The original images were collected by a road inspection vehicle driving across bridges on national and provincial highways in Jiangxi Province, China. The images on surface defects of cement concrete bridges were selected, and divided into a training set and a test set, and preprocessed through morphology-based weight adaptive denoising. To verify its performance, the improved VGG-16 was compared with traditional shallow neural networks (NNs) like the backpropagation neural network (BPNN), support vector machine (SVM), and deep CNNs like AlexNet, GoogLeNet, and ResNet on the same sample dataset of surface defects on cement concrete bridges. Judging by mean detection accuracy and top-5 accuracy, our model outperformed all the contrastive methods, and accurately differentiated between images with seven classes of defects such as normal, cracks, fracturing, plate fracturing, corner rupturing, edge/corner exfoliation, skeleton exposure, and repairs. The results indicate that our model can effectively extract the multi-layer features from surface defect images, which highlights the edges and textures. The research findings shed important new light on the detection of surface defects and classification of defect images. Full article

The acceptance sampling plan and process capability index (PCI) are critical decision tools for quality control. Recently, numerous research papers have examined the acceptance sampling plan in combination with the PCI. However, most of these papers have not considered the aspect of rectifying inspections. In this paper, we propose a quality cost model of repetitive sampling to develop a rectifying acceptance sampling plan based on the one-sided PCI. This proposed model minimizes the total quality cost (TQC) of sentencing one lot, including inspection cost, internal failure cost, and external failure cost. In addition, sensitivity analysis is conducted to investigate the behavior of relevant parameters against TQC. To demonstrate the advantages of the proposed methodology, a comparison is implemented with the existing rectifying sampling plan in terms of TQC and average outgoing quality limit. This comparison reveals that our proposed methodology exhibits superior performance. Full article

The Northern Territory Top End Health Service, Medical Entomology Section and the City of Darwin council carry out a joint Mosquito Engineering Program targeting the rectification of mosquito breeding sites in the City of Darwin, Northern Territory, Australia. In 2005, an investigation into potential subterranean stormwater breeding sites in the City of Darwin commenced, specifically targeting roadside stormwater side entry pits. There were 79 side entry pits randomly investigated for mosquito breeding in the Darwin suburbs of Nightcliff and Rapid Creek, with 69.6% of the pits containing water holding sumps, and 45.6% of those water holding sumps breeding endemic mosquitoes. Culex quinquefasciatus was the most common mosquito collected, accounting for 73% of all mosquito identifications, with the potential vector mosquito Aedes notoscriptus also recovered from a small number of sumps. The sumps were also considered potential dry season maintenance breeding sites for important exotic Aedes mosquitoes such as Aedes aegypti and Aedes albopictus, which are potential vectors of dengue, chickungunya and Zika virus. Overall, 1229 side entry pits were inspected in ten Darwin suburbs from 2005 to 2008, with 180 water holding sumps identified and rectified by concrete filling. Full article

Railway tunnel 3D clearance inspection is critical to guaranteeing railway operation safety. However, it is a challenge to inspect railway tunnel 3D clearance using a vision system, because both the spatial range and field of view (FOV) of such measurements are quite large. This paper summarizes our work on dynamic railway tunnel 3D clearance inspection based on a multi-camera and structured-light vision system (MSVS). First, the configuration of the MSVS is described. Then, the global calibration for the MSVS is discussed in detail. The onboard vision system is mounted on a dedicated vehicle and is expected to suffer from multiple degrees of freedom vibrations brought about by the running vehicle. Any small vibration can result in substantial measurement errors. In order to overcome this problem, a vehicle motion deviation rectifying method is investigated. Experiments using the vision inspection system are conducted with satisfactory online measurement results. Full article

The need for defect reporting is becoming increasingly difficult to ignore at handover of new residential buildings. A general review in defect studies has consistently shown that newly built properties can be found to have a significant number of defects. Very often the responsibility for rectifying these common defects is borne by the new homeowner even though house developers are liable. In the current study, survey data is obtained from 216 recent home purchasers/owners across New Zealand urban cities. The intent of the investigation is to show that opportunities exist for defect reporting that will act as a mechanism to measure performance and thus improve the quality of finished construction products in New Zealand. The study found that a significant number (81%) of the participants were involved in the construction of their homes and could influence quality performance if they were proactive enough. The results show that (64.7%) did not engage the service of independent building inspectors for defect reporting on their new homes. Seventy-four percent now agree that independent building inspection was important in hindsight. The study findings are in line with literature on defects and the poor use of defect reporting in new residential buildings. The current challenge for defect rectification by house developers after handover is real and this could increase the confidence that new home owners can have in their developers. Defect reporting could confer benefits to new residential building quality in New Zealand and should be embraced as part of a wider best practice. Full article