17362 research outputs found
Sort by
An annotated image dataset of urban insects for the development of computer vision and deep learning models with detection tasks
No full textA large image dataset with the aim of developing an insect recognition algorithm like YOLO. The dataset contains more than 25,000 annotations on the taxonomy of urban insects according to their order and the localization of the insect (as a bounding box) on a scanned image. This annotated image dataset of flying insects was collected using UV light traps placed in food warehouses, manufacturers and grocery stores in urban environments. The traps, equipped with UVA lamps (365 nm), captured a variety of insect species on sticky cards over 7–10 days. The sticky traps with all captured insects were used to create high-resolution scanned images (1200 dpi, 48-bit colour), with the resolution preserving fine morphological details of the insect, such as the antenna. To annotate the dataset for computer vision and deep learning models with detection tasks, annotation was performed using CVAT, with bounding boxes labelled by entomology experts at the order level. The dataset was intended to serve as a dataset for computer scientists or entomologists to compare the performance of deep learning models that can be used to build an automatic detection system for urban insect diversity or pest control studies
MOF-derived carbon nanotube/vertical graphene composite: A binder-free electrode for high-performance supercapacitors with aqueous redox electrolyte
No full textVertical graphene (VG), with its unique 3D structure and exceptional properties, is considered an excellent electrode material for supercapacitors. However, the strong hydrophobicity of VG significantly limits its performance in aqueous supercapacitors. In this study, we successfully construct a hybrid carbon nanotubes-vertical graphene-carbon cloth (CNT-VG-CC) structure by converting metal-organic frameworks (MOFs) into carbon nanotubes (CNTs) on the VG surface. This hybrid structure is then employed to fabricate a novel binder-free electrode for hybrid supercapacitors, which operate in a KOH electrolyte with redox additives, K3Fe(CN)6/K4Fe(CN)6. The resulting CNT-VG-CC hybrid structure not only exhibits superhydrophilicity but also offers a large effective specific surface area, enhancing electrolyte contact. The CNT-VG-CC electrode demonstrates an ultrahigh areal capacitance of 1526.5 mF/cm2 at 10 mA/cm2, nearly 100 % coulombic efficiency, and exceptional cycling stability, retaining 100 % of its capacitance after 10,000 charge-discharge cycles. The assembled supercapacitor device delivers a high energy density of 179.7 μW h/cm2 at a power density of 8000.0 μW/cm2. Remarkably, after 10000 cycles, it retains 118.9 % of its initial areal capacitance and maintains a coulombic efficiency of 90.7 %. These impressive performances can be attributed to the unique CNT-VG-CC hybrid structure, its super-hydrophilic property, high electronic conductivity, lager effective specific surface area, and synergistic interaction between the redox electrolyte and the CNT-VG-CC electrode
Fiber treatment impact on the thermal behavior of biomass/palm-fibers polylactic-acid hybrid biocomposites
No full textThe purpose of this investigation was to develop 100 % natural biocomposites from biopolymers such as polylactic acid (PLA) based on treated fibers and biochar (B) derived from the Washingtonia filifera (WF) plant. The dynamic mechanical characteristics, coefficient of thermal expansion, and thermal stability (TS) of biocomposites were studied using dynamic mechanical and thermomechanical analyses. Increasing the fiber treatment period caused the melting temperature of the biocomposites to decrease. The PLA-BWF72 composite demonstrated better TS compared to the others. Compared to untreated hybrid biocomposites, these composites exhibit enhanced TS and resistance following treatment with sodium bicarbonate. The dynamic mechanical analysis revealed that PLA-BWF72 hybrid biocomposites (2621.987 MPa) had a significantly higher storage modulus (SM) than the biocomposites produced. However, PLA-BWF24 hybrid biocomposites showed the lowest SM (2299.174 MPa), indicating a low level of stiffness. Cole-Cole plots of the hybrid biocomposites developed revealed the presence of imperfect semicircles, indicating their heterogeneity
Fatal e-cigarette or vaping product use-associated lung injury secondary to streptococcus constellatus empyema
No full textStreptococcus constellatus empyema in e-cigarette or vaping product use-associated lung injury can be fatal. Empyema as a presentation of e-cigarette or vaping product use-associated lung injury (EVALI) is rarely reported. Streptococcus constellatus is a known causative pathogen in empyema, but is usually associated with the elderly and immunocompromised states. A 13-year-old male with a history of active e-cigarette use presented to our centre with right lung empyema, which was complicated by suspected bronchopleural fistula. Despite a chest drain, antibiotics, inotropic and ventilatory support, he eventually succumbed to his illness. This case highlights a rarely reported severe complication of EVALI. Clinicians should be alert to the history of e-cigarette use in patients presenting with unexplained empyema
User identification and verification based on auditory evoked potentials using CNN
No full textIn recent years, researchers have focused on the biometric applications of bioelectrical signals, particularly electroencephalograms (EEG), to enhance information security. Using EEG as a biometric offers advantages that cannot be forgotten or forged. One approach to utilizing EEG signals for biometric purposes involves recording auditory evoked potentials (AEP). AEPs are electrical potentials that arise in response to auditory stimulation in the cerebral cortex. These signals are stimulus-dependent and can vary with the auditory stimulus, allowing these signals to be employed even if the registered signal was compromised. In this paper, discriminative features are extracted and classified using convolutional neural networks. A dataset recorded from 20 users using auditory stimulation is analyzed. The reported results demonstrate a classification accuracy of 98.99% in identification mode and an equal error rate of 1.18% in verification mode. These outcomes showcase the proposed method’s high accuracy, marking an improvement over existing methods. Furthermore, the system’s practicality is enhanced by utilizing fewer channels, and its performance is assessed by reducing the number of channels
The development of science laboratory safety educational comic for secondary school students in Malaysia
No full textThe traditional method of teaching science laboratory safety often involves lengthy, complex manuals that are difficult for students to understand. Despite the existence of the School Science Laboratory Management and Safety (SLMS) guidebook, students’ knowledge and awareness of lab safety remain low. Research shows that comics can enhance students’ interest and understanding of safety protocols. This study aimed to develop a science laboratory safety educational comic for secondary school students to raise awareness about lab safety. The educational comic was developed following the analysis, design, development, implement, and evaluate (ADDIE) model, with this study focusing on the first three stages. A needs analysis involved questionnaires given to 131 students in Kuala Lumpur and Kedah, revealing a moderate awareness of lab regulations (average score: 3.53) and strong support for developing a safety educational comic (average score: 3.90). The educational comic, designed by six experts from university academic and science teachers, and illustrated by a professional illustrator, covers topics such as laboratory preparation, waste disposal, laboratory accidents, and fire extinguishers. Validation was performed by experts in materials aspect (2 education academic experts), learning aspect (2 science teachers), and media design aspect (2 art and design academic expert), with scores of 4.80, 4.85, and 4.65, respectively, indicating high qualifications. In conclusion, the science laboratory safety educational comic was found to be valid and highly effective in meeting educational standards for lab safety
Skink stargazing: On the status and taxonomy of the Sphenomorphus stellatus complex (Boulenger, 1900) in Borneo, with the description of a new species (Squamata: Scincidae)
No full textSphenomorphus stellatus was once thought to be a wide-ranging species distributed across many parts of Indochina, the Thai-Malay Peninsula and Borneo. However, recent integrative taxonomic studies on the species have revealed its underestimated diversity and new species previously masquerading under this nomen in mainland Southeast Asia. This prompted the re-examination of a specimen identified as S. stellatus from Borneo, and morphological analyses revealed that it is not conspecific with topotypic specimens of the species from Peninsular Malaysia or other congeners of the complex, and is therefore described herein as new. Sphenomorphus striatafaucium sp. nov. can be differentiated from its congeners by a combination of known adult SVL of 57.4 mm; frontal scale undivided; prefrontals in broad contact; four supraoculars; two loreals; eight superciliaries; no enlarged, clear, central window in lower eyelid; seven or eight supralabials; six infralabials; three pairs of chinshields; 24 midbody scale rows; 62 paravertebrals, not wider than other dorsals; 66 ventrals; two enlarged precloacals; 19 subdigital lamellae on fourth toe; and bold and distinct striped pattern on the chin and throat. The discovery of S. striatafaucium sp. nov. increases the number of endemic Sphenomorphus skinks in Borneo to 13 species. The discovery underscores the underestimated diversity of reptile and amphibian species in Borneo, and the need for more concerted efforts to conduct fieldwork and systematic studies to not only improve the understanding of their relationships and ecologies, but also better protect their habitats that are increasingly threatened by anthropogenic pressures. This study also highlights the value of well-curated natural history collections in museums that still harbour new species awaiting discovery
The relationship between students' computer system skills and their readiness for using virtualization technology in learning at vocational college
No full textThis quantitative survey type study aims to examine the relationship between students' computer system skills and students' willingness to use virtualization technology in the learning process at the Vocational College in Tawau, Sabah. A total of 92 students in the vocational certificate and vocational diploma programs were selected as study respondents. Data was collected through the use of questionnaires and analyzed using descriptive statistics and inferential statistics using Statistical Software for Social Sciences (SPSS). The results of descriptive statistical analysis show that both students' computer system skills and their willingness to use virtualization technology are at a high level. Pearson Moment correlation analysis shows that there is a strong and significant relationship between the two variables. These results show that the higher the computer system skills and readiness of students, the more likely they will use virtualization technology in the learning process in vocational colleges. This study provides strong evidence that there is a significant positive relationship between students' computer system skills, their readiness, and the use of virtualization technology in the context of learning in a vocational college. Therefore, it is important for educational institutions to focus on the development of students' computer system skills and strengthen their readiness for the use of virtualization technology to increase the effectiveness of the learning process in this digital era
Assessment of the fitness test 10-metre underwater run among military cadet officers
No full textThis study assesses the reliability of the 10-metre underwater run as a fitness test among military cadet officers, focusing on its ability to evaluate anaerobic capacity, breath-hold endurance, and adaptability to aquatic environments. As military operations increasingly require performance in water-based settings, traditional land-focused assessments may not fully capture operational readiness. The 10-metre underwater run challenges cadets to hold their breath while sprinting underwater, engaging both aerobic and anaerobic energy systems. Using a quantitative test-retest approach, 30 cadet officers from the National Defence University of Malaysia completed the test twice, with a seven-day interval. Pearson’s correlation coefficient was used to determine consistency between test sessions, resulting in a high correlation of 0.810 (p < 0.001), indicating strong reliability. This reliability supports the test as a robust tool for assessing underwater sprint capacity and breath-hold endurance. The results align with previous findings on the importance of repeatability in operational fitness tests, suggesting that the 10-metre underwater run could serve as a valuable tool in military training for evaluating cadet readiness in aquatic scenarios. Future studies could expand the sample size and explore additional variables, enhancing the understanding of performance in underwater environments and strengthening military training frameworks for diverse operational demands
Humanism enhancing student motivation in history learning through ai-driven e-cooperative and e-collaborative methods in rural education
No full textThis research explores the influence of AI-based e-cooperative and e-collaborative approaches on boosting student motivation in history education in rural areas. Through the integration of primary data obtained from student surveys and teacher interviews, quantitative indicators of academic achievement, and qualitative insights from case studies, the study emphasizes how AI-driven platforms customize learning experiences, encourage peer collaboration, and enhance engagement. Results show that collaboration enhanced by AI boosts participation levels, enhances historical awareness, and fosters a sense of togetherness among students in distant regions. The research highlights the ability of AI to revolutionize history education by enhancing interactivity, accessibility, and engagement for students in disadvantaged rural areas