DR-NTU (Digital Repository of NTU)
Not a member yet
116018 research outputs found
Sort by
Zinc finger protein2 suppresses funiculus lignification to ensure seed loading efficiency in Arabidopsis
No full textThe plant funiculus anchors the developing seed to the placenta within the inner dorsal pod strands of the silique wall and directly transports nutrients to the seeds. The lignified vasculature critically supports nutrient transport through the funiculus. However, molecular mechanisms underlying lignified secondary cell wall (SCW) biosynthesis in the funiculus remain elusive. Here, we show that the transcription factor ZINC FINGER PROTEIN2 (ZFP2) represses SCW formation in the cortex cells that surround the vasculature. This function is essential for efficient nutrient loading into the seeds. Notably, ZFP2 directly acts on the SCW transcription factor NAC SECONDARY WALL THICKENING PROMOTING FACTOR1 (NST1) to repress cortex cell lignification, providing a mechanism of how SCW biosynthesis is restricted to the vasculature of the funiculus to ensure proper seed loading in Arabidopsis.Ministry of Education (MOE)National Research Foundation (NRF)Submitted/Accepted versionThis work was supported by the Ministry of Education (MOE) of Singapore Tier 2 (grant nos. MOE-T2EP30123-0001 and MOE-T2EP30220-0011 to W.M.), MOE of Singapore Tier 1 (grant no. RG32/23 to W.M.), Villum Investigator (grant no. 25915 to S.P.), DNRF Chair (grant no. DNRF155 to S.P.), Novo Nordisk Laureate (grant no. NNF19OC0056076 to S.P.), Novo Nordisk Emerging Investigator (grant no. NNF20OC0060564 to S.P.), Novo Nordisk Data Science (grant no. NNF0068884 to S.P.), Independent Research Fund Denmark (grant no. 2102-00258B to D.X.), Innovation Fund Denmark (grant no. 3141-00040B to D.X.), Novo Nordisk Foundation (grant no. NNF23OC0085289 to D.X.), MOE of Singapore Tier 2 (grant no. MOE-T2EP30122-0021 to Y.M.), Singapore National Research Foundation Investigatorship (grant no. NRF-NRFI08-2022-0012 to Y.M.), and MOE of Singapore Tier 2 (grant no. MOE-T2EP30122-0017 to M.M.). L.B. is supported by the EMBO postdoctoral fellowship ALTF 37-2022
Collaborative visual localization system for search and rescue operations
No full textMulti-agent Visual SLAM is an important topic in computer vision, due to its appli-
cations in real-life scenarios, especially in search and rescue operations. This study
has a three-pronged approach to tackle this issue, and it combines experimentation and
visualisation to explore Visual SLAM. It evaluates the effectiveness of Visual SLAM
algorithms, specifically ORB-SLAM3 and COVINS. It relies on these algorithms to
determine the feasibility of multiple agents contributing to form a singular unified map
of an environment. Using real-world datasets such as the EuroC dataset and webcam
live feeds, its applicability in multi-agent conditions have been analysed. Key aspects
of Visual SLAM such as feature extraction, feature matching, and loop closure are
also analysed in-depth in this paper. This study highlights the strengths of computer
vision in real-life applications, as the garnered findings were able to provide insights
on collaborative mapping and map-merging.Bachelor's degre
Prime minister Anwar Ibrahim at mid-term
No full textAnwar Ibrahim is on track to complete his first full term as prime minister. While he presides over a stellar economy, racial politics continue to cast a shadow over his administration.Published versio
Remolding the interface stability for practical aqueous Zn/I2 batteries via sulfonic acid-rich electrolyte and separator design
No full textThe electrolyte-electrode interface plays a crucial role in aqueous Zn/I2 battery and is largely determined by the properties of electrolyte and separator. Here, the synergistic effect of sulfonic acid-rich electrolyte additive and separator impacts the interface stability of Zn/I2 batteries is comprehensively investigated using operando synchrotron-based Fourier-transform infrared spectroscopy, cryo-electron microscopy, and in situ spectroscopy. As a case study, a cost-effective additive known as lignosulfonic acid sodium (LAS) and a flexible sulfonated polyether sulfone membrane are employed to facilitate the formation of a stable solid electrolyte interface (SEI) on the Zn anode and effectively suppress the shuttle effect. The chemisorption of LAS on Zn, its interaction with Zn2+, and the impact on the Zn desolvation process are systematically investigated through both theoretical simulations and operando measurements. Furthermore, the formation of an in situ SEI consisting of ZnS and ZnF2 is identified, which facilitates the uniform nucleation and planar plating of Zn(002), while effectively suppressing detrimental side reactions. Additionally, visualization experiments and in situ spectroscopy confirm that R-SO3- groups effectively impede the shuttle process of I3-/I5- anions through electrostatic repulsion. This work provides valuable insights for designing robust electrolyte interfaces for high-performance aqueous Zn/I2 batteries.Submitted/Accepted versionThe authors thank the financial support from the National Key R & D Project (2024YFE0101100), the National Natural Science Foundation of China (52172173, 52302205), the Excellent Research and Innovation Team Project of Anhui Province (2022AH010001), and the Natural Science Foundation of Anhui Province for Distinguished Young Scholars (2108085J25)
君绍散文中的自然书写 = Natural writing in the prose of Jun Shao
No full text在新加坡华文文学中，君绍是一位不可忽视的自然书写作家。他的散文反映了他对大自然的热爱，也记录了新加坡的自然景观和历史。然而，现有的学术研究对君绍的自然书写缺乏深入探讨，长期未能引起足够的关注。因此，本论文将会补充这一空白，探讨君绍散文中的自然书写。
论文将从生态批评的角度出发，分析君绍如何在散文中体现生态伦理观。此外，借助人文主义地理学中的地方论述，考察田园对君绍在情感与身份认同上的影响。最后，论文将分析君绍散文如何承载南洋色彩及历史记忆。通过分析新华作家君绍的散文作品，进一步拓展自然文学的研究边界。
In Singapore Chinese literature, Jun Shao is a significant nature writer whose essays reflect his love for nature while documenting Singapore’s natural landscapes and history. However, current academic research has overlooked Jun Shao’s nature writing, failing to give it adequate attention. Therefore, this paper seeks to fill this gap by examining the nature writing in Jun Shao’s essays.
Adopting an ecocritical perspective, this study examines how Jun Shao’s prose reflect his ecological ethics. Drawing on the concept of place discourse from humanistic geography, it also explores how his hometown’s landscape shapes his emotions and identity. Furthermore, the analysis highlights how his prose capture local characteristics and historical memory. Through this in-depth exploration, the research seeks to broaden the scope of nature literature studies.Bachelor's degre
Fiction and emotion: a critical evaluation of ontological and epistemic responses to the paradox of fiction
No full textThe Paradox of Fiction poses a major philosophical problem in modern philosophy. It reveals the apparent contradiction between our emotional involvement in fictional narratives and our knowledge that they are not real. This paper critically evaluates two primary responses to the paradox, the ontological response and the epistemic response, ultimately defending the latter. By rooting emotional investment in the state of imagination rather than belief, the epistemic response weakens the paradox by challenging its core assumption, that belief is a necessary condition for the development of genuine emotional responses. This paper builds on that and demonstrates that fiction can articulate meaningful and non-traditional truths—existential, emotional and narrative—that enhance our understanding of ourselves and others. While much could be discussed regarding the status of these non-propositional truths, the epistemic view offers a more realistic and satisfying explanation of how we make sense of fiction.Bachelor's degre
Dual type-II colloidal quantum wells for efficient nonlinear optical limiting
No full textColloidal II-VI nanocrystals have garnered significant research attention in nonlinear optical applications due to their low-cost synthesis, photophysical tunability, and ease of device integration. Herein, we report that dual-type II CdSe/CdTe/CdSe colloidal quantum wells (CQWs) with core/crown/crown structures achieve remarkable nonlinear optical limiting capabilities driven by an exceptionally large nonlinear absorption coefficient. Open aperture Z-scan reveals that these dual-type II CQWs exhibit a third-order nonlinear absorption coefficient of 33.1 cm/GW and an ultralow optical limiting threshold (0.71 GW/cm2), which is superior to that of any other reported colloidal semiconductor nanocrystals while also being comparable to existing two-dimensional (2D) dichalcogenide sheets. Photophysical analysis indicates that such a remarkable nonlinear optical performance in dual-type II CQWs can be primarily ascribed to the efficient excited state absorption (i.e., the sequential two-photon absorption), which benefits from the ultrafast and uniform formation of charge separation states in the dual type-II heterostructures.Agency for Science, Technology and Research (A*STAR)This work was supported by the National Key Research and Development Program (2023YFB3611100), the National Natural Science Foundation of China under Grant 12304032; Singapore Agency for Science, Technology and Research (A*STAR) MTC program (Grant No. M21J9b0085). Partial support was also provided by TUBITAK 121C266, and 20AG001. H.V.D. would like to acknowledge the support from the TUBA and TUBITAK 2247-A National Leader Researchers Program
A lightweight approach to crowd density estimation: leveraging network pruning for model compression
No full textCrowd density estimation is crucial for public safety, event management, and urban planning, particularly in high-density environments such as sports events, concerts, and transportation hubs. Traditional methods, including manual counting and statistical modeling, are often inefficient and error prone. Advancements in deep learning and convolutional neural networks (CNNs) have significantly improved accuracy, with models like the Pyramid Scale Network (PSNet) achieving state-of-the-art performance. However, the high computational demands of these models present challenges for real-time deployment on resource-constrained devices. This study addresses these limitations by applying model compression techniques, specifically network pruning, to optimize PSNet for efficient crowd density estimation. Experiments conducted using the ShanghaiTech dataset demonstrated that the compressed PSNet models achieved reductions in model size and inference time. Remarkably, most experimental results indicated an improvement in model accuracy after pruning, while other results showed minimal trade-offs in accuracy. These findings contribute to advancing the feasibility of deploying compressed models in resource-constrained environments, offering a promising lightweight solution for real-time crowd density estimation.Bachelor's degre
Synergistic impact of temperature and pore saturation on corrosion in carbonated reinforced concrete
No full textCarbonation-induced corrosion in reinforced concrete (RC) structures significantly threatens their long-term durability, particularly in lower-strength concrete typical of 1960s-1970s buildings in high temperature regions like Singapore. Despite extensive research on individual factors affecting corrosion, the synergistic effects of temperature and pore saturation remain poorly understood in warmer climates. Current durability assessment tends to overestimate the service life of carbonated RC structures in tropical climates due to incomplete understanding of these combined effects. This study investigates the temperature-pore saturation relationship through systematic experimental variation (0–40 °C, 60–100 % pore saturation) using electrochemical and non-destructive measurements, aiming to develop accurate predictive tools for RC maintenance in tropical climates. Results revealed a critical phenomenon: minimum pore saturation threshold for severe corrosion decreases significantly with increasing temperature. For concrete with a water-to-cement ratio of 0.65, severe corrosion occurred at substantially lower pore saturation levels at higher temperatures, with even lower thresholds for more porous concrete. To translate these findings into practical tools, multiple linear regression analysis yielded predictive equations relating temperature and pore saturation effects to corrosion rates. Analysis also revealed that temperature-dependent corrosion kinetics, rather than concrete drying, predominantly drive the corrosion process at higher temperatures, explaining substantial corrosion activity even at lower pore saturation levels. These findings prompt a fundamental reassessment of models that rely on temperature-independent pore saturation thresholds, as they significantly underestimate corrosion risks in hot climates.Ministry of National Development (MND)National Research Foundation (NRF)Published versionThis research is partially supported by the National Research Foundation, Singapore, and Ministry of National Development, Singapore, under its Cities of Tomorrow R&D Programme (CoT Award No. COT-V2-2019-1). Any opinions, findings and conclusions expressed in this material are those of the authors and do not reflect the views of National Research Foundation, Singapore and Ministry of National Development, Singapore
Algorithmic fast-charging for high energy-density Li-ion batteries
No full textThe burgeoning demand for rechargeable lithium-ion (Li-ion) batteries in industries such as consumer electronics and electric vehicles is clear and present. Advancement of battery technology in keeping with these fast-moving industries will mean improving on its strengths, while minimizing the drawbacks. Non-linear Voltammetry (NLV) is a fast-charging algorithm under development by KVI Battery aiming to best existing methods in charging speed, without putting heavy strain on the battery and reducing its lifespan. It achieves this by balancing the battery state, time requirements and the required charge capacity.
Our project compares the NLV algorithm against the industry-standard Constant Current-Constant Voltage (CCCV) method, evaluating factors such as degradation and efficacy in terms of charge speed and capacity attained, to determine the feasibility of NLV. Prior analysis was conducted on high power density cells with lower capacities. High-capacity cells have a high power density, and as such determining NLV charge parameters for these cells must be done in turn.
Utilising the NLV software and algorithm, we apply charge profiles to high energy battery cells and perform a systematic analysis of the charge performance results in response to changes in the NLV configuration. We use this understanding to extend the software’s functionality allowing for added flexibility in the charge parameters, maintaining charging efficacy and minimising capacity loss for different battery types.
Due to the characteristics and electrochemistry of high-density cells, the charge temperature, rest periods and use of a saturation charge are important parameters. Upon the conclusion of long-term parallel testing comparing these parameters to control values, there is evidence to show that high-density cells are compatible with the NLV charging algorithm, and a sample configuration for the battery can be derived.Bachelor's degre