Search results

The paper proposes a privacy-preserving artificial intelligence-enabled video surveillance technology to monitor social distancing in public spaces.

The paper proposes a new Responsible Artificial Intelligence Implementation Framework to guide the proposed solution's design and development. It defines responsible artificial intelligence criteria that the solution needs to meet and provides checklists to enforce the criteria throughout the process. To preserve data privacy, the proposed system incorporates a federated learning approach to allow computation performed on edge devices to limit sensitive and identifiable data movement and eliminate the dependency of cloud computing at a central server.

The proposed system is evaluated through a case study of monitoring social distancing at an airport. The results discuss how the system can fully address the case study's requirements in terms of its reliability, its usefulness when deployed to the airport's cameras, and its compliance with responsible artificial intelligence.

The paper makes three contributions. First, it proposes a real-time social distancing breach detection system on edge that extends from a combination of cutting-edge people detection and tracking algorithms to achieve robust performance. Second, it proposes a design approach to develop responsible artificial intelligence in video surveillance contexts. Third, it presents results and discussion from a comprehensive evaluation in the context of a case study at an airport to demonstrate the proposed system's robust performance and practical usefulness.

The purpose of this research is to achieve multi-task autonomous driving by adjusting the network architecture of the model. Meanwhile, after achieving multi-task autonomous driving, the authors found that the trained neural network model performs poorly in untrained scenarios. Therefore, the authors proposed to improve the transfer efficiency of the model for new scenarios through transfer learning.

First, the authors achieved multi-task autonomous driving by training a model combining convolutional neural network and different structured long short-term memory (LSTM) layers. Second, the authors achieved fast transfer of neural network models in new scenarios by cross-model transfer learning. Finally, the authors combined data collection and data labeling to improve the efficiency of deep learning. Furthermore, the authors verified that the model has good robustness through light and shadow test.

This research achieved road tracking, real-time acceleration–deceleration, obstacle avoidance and left/right sign recognition. The model proposed by the authors (UniBiCLSTM) outperforms the existing models tested with model cars in terms of autonomous driving performance. Furthermore, the CMTL-UniBiCL-RL model trained by the authors through cross-model transfer learning improves the efficiency of model adaptation to new scenarios. Meanwhile, this research proposed an automatic data annotation method, which can save 1/4 of the time for deep learning.

This research provided novel solutions in the achievement of multi-task autonomous driving and neural network model scenario for transfer learning. The experiment was achieved on a single camera with an embedded chip and a scale model car, which is expected to simplify the hardware for autonomous driving.

The paper aims to solve the problem of personnel intrusion identification within the limits of high-speed railways. It adopts the fusion method of millimeter wave radar and camera to improve the accuracy of object recognition in dark and harsh weather conditions.

This paper adopts the fusion strategy of radar and camera linkage to achieve focus amplification of long-distance targets and solves the problem of low illumination by laser light filling of the focus point. In order to improve the recognition effect, this paper adopts the YOLOv8 algorithm for multi-scale target recognition. In addition, for the image distortion caused by bad weather, this paper proposes a linkage and tracking fusion strategy to output the correct alarm results.

Simulated intrusion tests show that the proposed method can effectively detect human intrusion within 0–200 m during the day and night in sunny weather and can achieve more than 80% recognition accuracy for extreme severe weather conditions.

(1) The authors propose a personnel intrusion monitoring scheme based on the fusion of millimeter wave radar and camera, achieving all-weather intrusion monitoring; (2) The authors propose a new multi-level fusion algorithm based on linkage and tracking to achieve intrusion target monitoring under adverse weather conditions; (3) The authors have conducted a large number of innovative simulation experiments to verify the effectiveness of the method proposed in this article.

To ensure the motion attitude and stable contact force of massage robot working on unknown human tissue environment, this study aims to propose a robotic system for autonomous massage path planning and stable interaction control.

First, back region extraction and acupoint recognition based on deep learning is proposed, which provides a basis for determining the working area and path points of the robot. Second, to realize the standard approach and movement trajectory of the expert massage, 3D reconstruction and path planning of the massage area are performed, and normal vectors are calculated to control the normal orientation of robot-end. Finally, to cope with the soft and hard changes of human tissue state and body movement, an adaptive force tracking control strategy is presented to compensate the uncertainty of environmental position and tissue hardness online.

Improved network model can accomplish the acupoint recognition task with a large accuracy and integrate the point cloud to generate massage trajectories adapted to the shape of the human body. Experimental results show that the adaptive force tracking control can obtain a relatively smooth force, and the error is basically within ± 0.2 N during the online experiment.

This paper incorporates deep learning, 3D reconstruction and impedance control, the robot can understand the shape features of the massage area and adapt its planning massage path to carry out a stable and safe force tracking control during dynamic robot–human contact.

Images taken from construction site interiors often suffer from low illumination and poor natural colors, which restrict their application for high-level site management purposes. The state-of-the-art low-light image enhancement method provides promising image enhancement results. However, they generally require a longer execution time to complete the enhancement. This study aims to develop a refined image enhancement approach to improve execution efficiency and performance accuracy.

To develop the refined illumination enhancement algorithm named enhanced illumination quality (EIQ), a quadratic expression was first added to the initial illumination map. Subsequently, an adjusted weight matrix was added to improve the smoothness of the illumination map. A coordinated descent optimization algorithm was then applied to minimize the processing time. Gamma correction was also applied to further enhance the illumination map. Finally, a frame comparing and averaging method was used to identify interior site progress.

The proposed refined approach took around 4.36–4.52 s to achieve the expected results while outperforming the current low-light image enhancement method. EIQ demonstrated a lower lightness-order error and provided higher object resolution in enhanced images. EIQ also has a higher structural similarity index and peak-signal-to-noise ratio, which indicated better image reconstruction performance.

The proposed approach provides an alternative to shorten the execution time, improve equalization of the illumination map and provide a better image reconstruction. The approach could be applied to low-light video enhancement tasks and other dark or poor jobsite images for object detection processes.

This study aims to investigate how trauma team members perceive technological equipment and tools in the trauma room (TR) environment and to identify how the technological equipment could be optimized in relation to the TR’s space.

A total of 21 focus group sessions were conducted with 69 trauma team members, all of whom worked in Level I TRs from six teaching hospitals in the USA.

The collected data was analyzed and categorized into three parent themes: imaging equipment, assistive devices and room features. The results of the study suggest that trauma team members place high importance on the availability and versatility of the technological equipment in the TR environment. Although CT scans are a usual procedure necessity in TRs, few facilities were optimized for easy access to CT-scanners for the TR. The implementation of cameras and screens was suggested as an improvement to accommodate situational awareness. Rapid sharing of data, such as imaging results, was highly sought after. Unorthodox approaches, such as the use of automatic doors, were associated with slowing down the course of actions.

This study provides health-care designers with the knowledge they need to make informed decisions when designing TRs. It will cover key considerations such as room layout, equipment selection, lighting and controls. Implementing the strategies will help minimize negative patient outcomes.

Level I TRs are a critical element of emergency departments and designing them correctly can significantly impact patient outcomes. However, designing a TR can be a complex process that requires careful consideration of various factors, including patient safety, workflow efficiency, equipment placement and infection control. This study suggests multiple considerations when designing TRs.

For the provision of smart library services to end users, tools of the Internet of Things (IoT) play a significant role. The study aims to discover the factors influencing the adoption of IoT in university libraries, investigate the impact of IoT on university library services and identify challenges to adopt IoT applications in university libraries.

A systematic literature review was carried out to address the objectives of the study. The 40 most relevant research papers published in the world’s leading digital databases were selected to conduct the study.

The findings illustrated that rapid growth in technology, perceived benefits, the networked world and the changing landscape of librarianship positively influenced the adoption of IoT in university libraries. The study also displayed that IoT supported library professionals to initiate smart library services, assisted in service efficiency, offered context-based library services, provided tracking facilities and delivered effective management of library systems. Results also revealed that a lack of technical infrastructure, security and privacy concerns, a lack of technological skills and unavailability of policy and strategic planning caused barriers to the successful adoption of IoT applications in university libraries.

The study has provided theoretical implications through a valuable addition to the current literature. It has also offered managerial implications for policymakers to construct productive policies for the implementation of IoT applications in university libraries for the attainment of fruitful outcomes. Finally, the study provides a baseline for understanding the adoption of IoT in academic libraries.

This study aims to investigate how message sidedness affects the impact of fake news posted on social media on consumers' emotional responses.

The study involves a face-tracking experiment in which 198 participants were exposed to different fake news messages concerning the COVID-19 vaccine. Specifically, participants were exposed to fake news using (1) a one-sided negative fake news message in which the message was entirely unfavorable and (2) a two-sided fake news message in which the negative message was mixed with favorable information. Noldus FaceReader 7, an automatic facial expression recognition system, was used to recognize participants' emotions as they read fake news. The authors sampled 17,450 observations of participants' emotional responses.

The results provide evidence of the significant influence of message sidedness on consumers' emotional valence and arousal. Specifically, two-sided fake news positively influences emotional valence, while one-sided fake news positively influences emotional arousal.

The current study demonstrates that research on fake news posted on social media may particularly benefit from insights regarding the potential but often overlooked importance of strategic design choices in fake news messages and their impact on consumers' emotional responses.

Construction progress monitoring (CPM) is considered a difficult and tedious task in construction projects, which focuses on identifying discrepancies between the as-built product and the as-planned design. Computer vision (CV) technology is applied to automate the CPM process. However, the synergy between the CV and CPM in literature and industry practice is lacking. This study aims to fulfil this research gap.

A Delphi qualitative approach was used in this study by conducting two interview rounds. The collected data was analysed using manual content analysis.

This study identified seven stages of CPM; data acquisition, information retrieval, verification, progress estimation and comparison, visualisation of the results and schedule updating. Factors such as higher accuracy in data, less labourious process, efficiency and near real-time access are some of the significant enablers in instigating CV for CPM. Major challenges identified were occlusions and lighting issues in the site images and lack of support from the management. The challenges can be easily overcome by implementing suitable strategies such as familiarisation of the workforce with CV technology and application of CV research for the construction industry to grow with the technology in line with other industries.

This study addresses the gap pertaining to the synergy between the CV in CPM literature and the industry practice. This research contributes by enabling the construction personnel to identify the shortcomings and the opportunities to apply automated technologies concerning each stage in the progress monitoring process.

The crane plays an essential role in modern construction sites as it supports numerous operations and activities on-site. Additionally, the crane produces a big amount of data that, if analyzed, could significantly affect productivity, progress monitoring and decision-making in construction projects. This paper aims to show the usability of crane data in tracking the progress of activities on-site.

This paper presents a pattern-based recognition method to detect concrete pouring activities on any concrete-based construction sites. A case study is presented to assess the methodology with a real-life example.

The analysis of the data helped build a theoretical pattern for concrete pouring activities and detect the different phases and progress of these activities. Accordingly, the data become useable to track progress and identify problems in concrete pouring activities.

The paper presents an example for construction practitioners and researcher about a practical and easy way to analyze the big data that comes from cranes and how it is used in tracking projects' progress. The current study focuses only on concrete pouring activities; future studies can include other types of activities and can utilize the data with other building methods to improve construction productivity.

The proposed approach is supposed to be simultaneously efficient in terms of concrete pouring detection as well as cost-effective. Construction practitioners could track concrete activities using an already-embedded monitoring device.

While several studies in the literature targeted the optimization of crane operations and of mitigating hazards through automation and sensing, the opportunity of using cranes as progress trackers is yet to be fully exploited.