Search

The advancement of deep learning has led to a growing demand and increase in research on computer vision–based construction site monitoring for improved safety and operational efficiency. These methods largely depend on supervised learning, requiring labeled data for optimal performance. However, when applied to new construction sites with varied environmental conditions, the effectiveness of these models is often compromised. Additionally, highly imbalanced object class distributions in the data sets, known as long-tailed objects, presents significant challenges during model training, considerably impacting performance. Recognizing this crucial gap in the field, this study proposes a novel approach to improve safety and operational efficiency at construction sites by leveraging a semisupervised learning approach for domain adaptation in long-tailed object detection. The method addresses the challenges of unbalanced class distribution and environmental variability in construction site monitoring, which often degrade the performance of computer vision models. By employing semisupervised learning, both labeled and unlabeled data are utilized in domain adaptation to unseen construction sites, considering both image-and object-level noise, thereby enhancing the model's adaptability to diverse working conditions. Based on the detection results, a risk scenario detection algorithm is also introduced for construction vehicles and workers. The efficacy of the proposed approach was validated through extensive experiments conducted on a comprehensive data set sourced from AIHub and CrowdHuman, in addition to actual self-labeled closed-circuit television (CCTV) data comprising 500 videos from construction sites' CCTV cameras. The evaluations revealed that the proposed method significantly outperforms conventional semisupervised learning by 9.76% on mean average precision for construction vehicle detection and by 3% for the worker detection model, paving the way for advanced construction site monitoring systems that ensure a safer and more efficient working environment. [ABSTRACT FROM AUTHOR]

Introduction: This study aims to understand the interactions of perception, effort, emotion, time and performance during the performance of multiple information tasks using Web information technologies. Method: Twenty volunteers from a university participated in this study. Questionnaires were used to obtain general background information and examine the research problem. Analysis: The quantitative data were analysed using the Statistical Package for the Social Sciences (SPSS). Results: The results of this study revealed that those with higher cognitive effort experienced more stress and frustration and higher temporal demand, resulting in lower performance, even though they spent more time finishing the information task. A high degree of temporal demand and negative emotional state led people to perform the task poorly even though they invested more mental effort on the task. Our feelings and emotions play a key role in our ability to deal with a situation where multiple information tasks need to be done within a certain time limit. Conclusions: The results of this study can be employed as a theoretical foundation for designing human-friendly, adaptive user interfaces, which function as intelligent and affective central mechanisms and help users prioritise, monitor and coordinate their needs/tasks/goals effectively and efficiently. This study introduces the emotional factor, which is a newly emerging dimension, in dynamic information seeking and retrieval contexts and enlightens the existing areas of human information interaction.

The material transport system, facilitated by motor proteins, plays a vital role in maintaining a non‐equilibrium cellular state. However, understanding the temporal coordination of motor protein activity requires an advanced imaging technique capable of measuring 3D angular displacement in real‐time. In this study, a Fourier transform‐based plasmonic dark‐field microscope has been developed using anisotropic nanoparticles, enabling the prolonged and simultaneous observation of endosomal lateral and rotational motion. A sequence of discontinuous 3D angular displacements has been observed during the pause and run phases of transport. Notably, a serially correlated temporal pattern in the intermittent rotational events has been demonstrated during the tug‐of‐war mechanism, indicating Markovian switching between the exploitational and explorational modes of motor protein exchange prior to resuming movement. Alterations in transition frequency and the exploitation‐to‐exploration ratio upon dynein inhibitor treatment highlight the relationship between disrupted motor coordination and reduced endosomal transport efficiency. Collectively, these results suggest the importance of orchestrated temporal motor protein patterns for efficient cellular transport. [ABSTRACT FROM AUTHOR]

Repeated pesticide residue detection in chili peppers in the Republic of Korea has become a serious health concern. Thus, monitoring domestically grown and imported chili peppers for pesticide residues is of great significance. Here, we investigated pesticide residues detected in imported and domestically grown chili peppers using global pesticide residue monitoring data. Our analysis involved organizing inspection and detection data from different sources. Global pesticide residue monitoring data for chili peppers revealed 139 pesticide types, 43,532 inspections, and 3966 detections (detection rate, 9.11%). Peppers from Mexico were sampled the most (39,927 inspections) and showed the highest number of detected cases (2998 cases). Globally, the top 10 most frequently detected pesticides were clothianidin, imidacloprid, thiamethoxam, chlorpyrifos, thiacloprid, metalaxyl, myclobutanil, azoxystrobin, carbendazim, and cyhalothrin, with detection rates in the range of 10.52–28.66%. Furthermore, domestic chili pepper pesticide residue monitoring revealed 73 pesticide types, 3535 inspections, and 332 detected cases (detection rate, 9.39%), and the top 10 most frequently detected pesticides were chlorfenapyr, tebuconazole, flonicamid, dinotefuran, boscalid, pyraclostrobin, fluxametamide, thiamethoxam, pyridaben, and azoxystrobin, with detection rates in the range of 13.89–32.58%. These findings may serve as fundamental data for safety management related to chili pepper pesticide residues in the Republic of Korea. [ABSTRACT FROM AUTHOR]