Search results

The swivel construction method is a specially designed process used to build bridges that cross rivers, valleys, railroads and other obstacles. To carry out this construction method safely, real-time monitoring of the bridge rotation process is required to ensure a smooth swivel operation without collisions. However, the traditional means of monitoring using Electronic Total Station tools cannot realize real-time monitoring, and monitoring using motion sensors or GPS is cumbersome to use.

This study proposes a monitoring method based on a series of computer vision (CV) technologies, which can monitor the rotation angle, velocity and inclination angle of the swivel construction in real-time. First, three proposed CV algorithms was developed in a laboratory environment. The experimental tests were carried out on a bridge scale model to select the outperformed algorithms for rotation, velocity and inclination monitor, respectively, as the final monitoring method in proposed method. Then, the selected method was implemented to monitor an actual bridge during its swivel construction to verify the applicability.

In the laboratory study, the monitoring data measured with the selected monitoring algorithms was compared with those measured by an Electronic Total Station and the errors in terms of rotation angle, velocity and inclination angle, were 0.040%, 0.040%, and −0.454%, respectively, thus validating the accuracy of the proposed method. In the pilot actual application, the method was shown to be feasible in a real construction application.

In a well-controlled laboratory the optimal algorithms for bridge swivel construction are identified and in an actual project the proposed method is verified. The proposed CV method is complementary to the use of Electronic Total Station tools, motion sensors, and GPS for safety monitoring of swivel construction of bridges. It also contributes to being a possible approach without data-driven model training. Its principal advantages are that it both provides real-time monitoring and is easy to deploy in real construction applications.

The purpose of this paper is to use vision stereo to simultaneously acquire image pairs under a normal environment. Then the methods of moving edges detection and moving target shifting are applied to reduce noise error in order to position a target efficiently. The target is then double confirmed via image merge and alignment. After positioning, the visual difference between the target and the image created by the stereo vision system is measured for alignment. Finally, the image depth of the target is calculated followed by real-time target tracking.

This study mainly applies Sobel image principle. In addition, moving edges detection and moving target shifting are also used to work with system multi-threading for improving image identification efficiency.

The results of the experiment suggest that real-time image tracking and positioning under a pre-set environment can be effectively improved. On the other hand, tracking and positioning are slightly affected under a normal environment. Errors of distance measurements occur because there is more noise existing.

This study mainly determines the movements and positioning of an object or a target via image. However, the stability of moving edges detection executed by the stereo vision system can be affected if the light sources in an environment are too strong or extreme.

So far the method of tracking and positioning a moving object has been applied to surveillance systems or the application which requires measuring and positioning under a normal environment. The method proposed by this study can also be used to construct a 3D environment.

The method proposed by this study can also be used to construct a 3D environment or tracking moving object to measure the distance.

This chapter introduces the approaches and methods employed in a four-country research project that resulted in the 2017 report The People in the Pictures: Vital perspectives on Save the Children’s image making. It presents and explores the ethical issues that emerged throughout the process of the research, particularly in relation to photo elicitation – the use of images (still and moving) within both interviews and focus groups. Interviews and focus groups took place in the UK, Jordan, Bangladesh, and Niger with a total of 202 research participants. The research involved sharing Save the Children content (fundraising materials, published reports, online news features, TV adverts, and short films) with research participants. Research participants included those featured in some of these visual communication materials (referred to as contributors), and other individuals within their communities (referred to as non-contributors). The following principles and decisions informed the research design: safe and ethical practice; inclusive, engaging and accessible approaches; the participation of children; prioritising first-hand accounts; no photography or filming; and the preparation of location- and language-specific resources for each interview and focus group. The main ethical issues to emerge during the design of the research related to predicting (and responding) to any potential negative impacts of the research on participants, particularly contributors, but also children. The researchers also experienced some unexpected ethical encounters, including visual materials causing some concern or distress. Additionally, assuring research participants’ anonymity led to the necessity of extra care when publishing the report and the use of images within that.

This paper aims to present a highly accessible and affordable tracking model for earthmoving operations in an attempt to overcome some of the limitations of current tracking models.

The proposed methodology involves four main processes: acquiring onsite terrestrial images, processing the images into 3D scaled cloud data, extracting volumetric measurements and crew productivity estimations from multiple point clouds using Delaunay triangulation and conducting earned value/schedule analysis and forecasting the remaining scope of work based on the estimated performance. For validation, the tracking model was compared with an observation-based tracking approach for a backfilling site. It was also used for tracking a coarse base aggregate inventory for a road construction project.

The presented model has proved to be a practical and accurate tracking approach that algorithmically estimates and forecasts all performance parameters from the captured data.

The proposed model is unique in extracting accurate volumetric measurements directly from multiple point clouds in a developed code using Delaunay triangulation instead of extracting them from textured models in modelling software which is neither automated nor time-effective. Furthermore, the presented model uses a self-calibration approach aiming to eliminate the pre-calibration procedure required before image capturing for each camera intended to be used. Thus, any worker onsite can directly capture the required images with an easily accessible camera (e.g. handheld camera or a smartphone) and can be sent to any processing device via e-mail, cloud-based storage or any communication application (e.g. WhatsApp).

This paper aims to propose a new deep learning technique to detect the type of material to improve automated construction quality monitoring.

A new data augmentation approach that has improved the model robustness against different illumination conditions and overfitting is proposed. This study uses data augmentation at test time and adds outlier samples to training set to prevent over-fitted network training. For data augmentation at test time, five segments are extracted from each sample image and fed to the network. For these images, the network outputting average values is used as the final prediction. Then, the proposed approach is evaluated on multiple deep networks used as material classifiers. The fully connected layers are removed from the end of the networks, and only convolutional layers are retained.

The proposed method is evaluated on recognizing 11 types of building materials which include 1,231 images taken from several construction sites. Each image resolution is 4,000 × 3,000. The images are captured with different illumination and camera positions. Different illumination conditions lead to trained networks that are more robust against various environmental conditions. Using VGG16 model, an accuracy of 97.35% is achieved outperforming existing approaches.

It is believed that the proposed method presents a new and robust tool for detecting and classifying different material types. The automated detection of material will aid to monitor the quality and see whether the right type of material has been used in the project based on contract specifications. In addition, the proposed model can be used as a guideline for performing quality control (QC) in construction projects based on project quality plan. It can also be used as an input for automated progress monitoring because the material type detection will provide a critical input for object detection.

Several studies have been conducted to perform quality management, but there are some issues that need to be addressed. In most previous studies, a very limited number of material types were examined. In addition, although some studies have reported high accuracy to detect material types (Bunrit et al., 2020), their accuracy is dramatically reduced when they are used to detect materials with similar texture and color. In this research, the authors propose a new method to solve the mentioned shortcomings.

This paper seeks to examine image retrieval within two different contexts: a monolingual context where the language of the query is the same as the indexing language and a multilingual context where the language of the query is different from the indexing language. The study also aims to compare two different approaches for the indexing of ordinary images representing common objects: traditional image indexing with the use of a controlled vocabulary and free image indexing using uncontrolled vocabulary.

This research uses three data collection methods. An analysis of the indexing terms was employed in order to examine the multiplicity of term types assigned to images. A simulation of the retrieval process involving a set of 30 images was performed with 60 participants. The quantification of the retrieval performance of each indexing approach was based on the usability measures, that is, effectiveness, efficiency and satisfaction of the user. Finally, a questionnaire was used to gather information on searcher satisfaction during and after the retrieval process.

The results of this research are twofold. The analysis of indexing terms associated with all the 3,950 images provides a comprehensive description of the characteristics of the four non‐combined indexing forms used for the study. Also, the retrieval simulation results offers information about the relative performance of the six indexing forms (combined and non‐combined) in terms of their effectiveness, efficiency (temporal and human) and the image searcher's satisfaction.

The findings of the study suggest that, in the near future, the information systems could benefit from allowing an increased coexistence of controlled vocabularies and uncontrolled vocabularies, resulting from collaborative image tagging, for example, and giving the users the possibility to dynamically participate in the image‐indexing process, in a more user‐centred way.

Histopathology biopsy imaging is currently the gold standard for the diagnosis of breast cancer in clinical practice. Pathologists examine the images at various magnifications to identify the type of tumor because if only one magnification is taken into account, the decision may not be accurate. This study explores the performance of transfer learning and late fusion to construct multi-scale ensembles that fuse different magnification-specific deep learning models for the binary classification of breast tumor slides.

Three pretrained deep learning techniques (DenseNet 201, MobileNet v2 and Inception v3) were used to classify breast tumor images over the four magnification factors of the Breast Cancer Histopathological Image Classification dataset (40×, 100×, 200× and 400×). To fuse the predictions of the models trained on different magnification factors, different aggregators were used, including weighted voting and seven meta-classifiers trained on slide predictions using class labels and the probabilities assigned to each class. The best cluster of the outperforming models was chosen using the Scott–Knott statistical test, and the top models were ranked using the Borda count voting system.

This study recommends the use of transfer learning and late fusion for histopathological breast cancer image classification by constructing multi-magnification ensembles because they perform better than models trained on each magnification separately.

The best multi-scale ensembles outperformed state-of-the-art integrated models and achieved an accuracy mean value of 98.82 per cent, precision of 98.46 per cent, recall of 100 per cent and F1-score of 99.20 per cent.

The purpose of this paper is to develop a novel and systematic framework for bridge inspection and management to improve the efficiency in current practice.

A new framework that implements camera-based unmanned aerial systems (UASs) with computer vision algorithms to collect and process inspection data, and Bridge Information Modeling (BrIM) to store and manage all related inspection information is proposed. An illustrative case study was performed using the proposed framework to test its feasibility and efficiency.

The test results of the proposed framework on an existing bridge verified that: high-resolution images captured by an UAS enable to visually identify different types of defects, and detect cracks automatically using computer vision algorithms, the use of BrIM enable assigning defect information on individual model elements, manage all bridge data in a single model across the bridge life cycle. The evaluation by bridge inspectors from 12 states across the USA demonstrated that all of the identified problems, except for being subjective, can be improved using the proposed framework.

The proposed framework enables to: collect and document accurate bridge inspection data, reduce the number of site visits and avoid data overload and facilitate a more efficient, cost-effective and safer bridge inspection process.

This paper contributes a novel and systematic framework for the collection and integration of inspection data for bridge inspection and management. The findings from the case study suggest that the proposed framework should help improve current bridge inspection and management practice. Furthermore, the difficulties experienced during the implementation are evaluated, which should be helpful for improving the efficiency and the degree of automation of the proposed framework further.

This paper aims to examine GCC companies' use of visual images to interplay modernity and globalism with tradition, Islam and local culture. The analysis aims to bring attention to the way that businesses in the GCC use visual images to engage with or influence debates in their societies concerning the tension between modernity, globalisation and traditional values in the Arab‐Islamic world.

The analysis is critical and discursive and based on a close reading of the visual images reported in the 2005 annual reports of companies listed on GCC stock markets.

The analysis suggests that GCC companies on many occasions used visual images to depict and represent the possibility of a successful profitable, modern and global business that is also sympathetic to tradition and operates within the framework of Islamic principles.

While visual images are increasingly used in companies' annual reports they have been largely ignored in accounting research. Furthermore, when this research manifests, it has been concerned with investigating Anglo‐American and Western contexts. This paper instead emphasises the significance of researching the use of visual images in a variety of contexts and locations. It critically and contextually explores the use of visual images in a largely unexplored, non‐Western and a significantly Islamic context.

This paper aims to explore how the Zaltman metaphor elicitation technique (ZMET) can be adopted in services marketing to provide deeper customer experience insights.

This paper explores how ZMET interviews, which use images selected by the participant to facilitate discussion, can be used by researchers. This paper draws upon a study of 24 student experiences at a UK university.

Adopting this qualitative method for services marketing can counter depth deficit when compared to other qualitative approaches, because it is participant led. However, the method requires competent interview skills and time for the interview and analysis. We find that ZMET has not been widely adopted in academia because of its commercial licenced use. The paper illustrates how to use the ZMET process step-by-step.

Findings are limited to student experiences. Further research is necessary to understand how researchers could use ZMET in other areas of services marketing.

This paper provides guidance to researchers on how to use ZMET as a methodological tool. ZMET facilitates a deeper understanding of service experiences through using participant chosen images and thus enabling researchers to uncover subconscious hidden perceptions that other methods may not find.

ZMET has been used commercially to gain market insights but has had limited application in service research. Existing studies fail to provide details of how ZMET can be used to access the consumer subconscious. This paper makes a methodological contribution by providing step-by-step guidance on how to apply ZMET to services marketing.