A new AI framework, AdvWT, uses realistic fading, cracks, and corrosion to test vulnerabilities in AI vision systems. It achieves near-perfect attack success rates on lightweight CNNs and transformer-based models, and can also be used to restore naturally damaged traffic signs.
Seoul National University researchers use probabilistic fracture mechanics to assess pipe rupture frequency in nuclear power plants, accounting for degradation mechanisms and individual parameters. The study highlights the importance of inspections in reducing rupture frequency, and the results can help engineers identify key factors g...
A new AI framework reduces computational effort needed to optimize solid oxide electrolysis cells, improving hydrogen production efficiency and thermal stability. The framework achieved 14% improvement in electrochemical performance index and 80% reduction in temperature differences compared to baseline conditions.
Researchers developed a multi-scale monitoring framework to detect ground settlement in urban metro corridors. The framework integrates InSAR, laser scanning, and ground penetrating radar to provide a comprehensive assessment of settlement and its underlying causes.
Seoul National University researchers have developed an off-stochiometric anode material that enables safe, ultra-fast charging of lithium-ion batteries. The new design strategy overcomes kinetic limitations and preserves the NASICON structure against irreversible damage.
Researchers developed an automated computer vision system to monitor long-term bridge damage using routine drone inspections. The framework enables the tracking of structural damage progression and measurement of severity, with a maximum error rate of 4.61% compared to manual measurements.
Researchers at Seoul National University have developed a novel approach using porous triply periodic minimal surface (TPMS) feet and deep reinforcement learning controller, which significantly reduces battery power consumption in quadruped robots. The solution reduces energy consumption by up to 6.2% while maintaining stable locomotion.
Researchers found perilla oil supplementation significantly prolonged platelet function and reduced inflammation in healthy smokers. The study also showed increased levels of omega-3 fatty acids and antioxidants in the body.
A new window technology shields buildings from EMP threats while maintaining transparency. The innovative design offers broadband EMP protection with high optical transparency, suitable for practical architectural applications.
The study demonstrates large NIR modulation using low-cost sodium electrolytes, comparable to lithium-based systems, with efficient heat-shielding performance. This breakthrough offers a practical solution for thermal regulation in diverse climate conditions.
A study by Seoul National University of Science & Technology found that expanding hydrogen fuel cell heavy-duty trucks could reduce carbon dioxide emissions by approximately 8.74 million tons. Public willingness to pay for this transition amounts to KRW 572.4 billion, far exceeding the prevailing carbon credit price.
VFF-Net applies label-wise noise labelling, cosine similarity-based contrastive loss, and layer grouping to improve image classification performance compared to conventional forward-forward networks. The algorithm reduces test errors on various datasets, enabling lighter and more brain-like training methods that make AI more sustainable.
Researchers at Seoul National University of Science and Technology have discovered fluorinated covalent organic polymers (FCOPs) that can remove 67.3% of beta-blockers like atenolol and metoprolol within the first minute. The FCOPs' strong adsorption performance is attributed to their rich structure, which enables multiple synergistic ...
Researchers developed an AI-based generative approach to discovering technology opportunities from patent maps using machine learning. The system translates patent vacancies into human-readable text, enabling the identification of untapped technologies and facilitating innovation forecasting.
Seoul National University researchers create highly stretchable, electrically conductive carbon nanotube-based nanocomposites using vat photopolymerization type 3D printing. The new material is optimized for smart health monitoring applications, enabling real-time pressure distribution detection.
A comprehensive safety assessment framework for liquid hydrogen storage systems in UAVs has been developed, addressing thermal performance and structural integrity challenges. The framework integrates multiple analyses, including thermal insulation, structural analysis, fatigue testing, and impact assessments.
A study by SeoulTech researchers found high levels of polycyclic aromatic hydrocarbons (PAHs) in common food items, including cooking oils and meats. The QuEChERS-GC-MS method revealed carcinogenic compounds with limits of detection ranging from 0.006 to 0.035 µg/kg.
Researchers at Seoul National University of Science and Technology developed a microelectrode with three-dimensional carbon nanotubes that efficiently conduct electricity while being soft like tissue. The arrays demonstrated stable insertion in brain tissues, precise recording of visual responses, and reduced inflammatory responses.
Researchers at Seoul National University of Science & Technology developed LWMalloc, a lightweight and high-performance dynamic memory allocator for resource-constrained environments. The new allocator outperforms ptmalloc by achieving up to 53% faster execution time and 23% lower memory usage.
The Seoul National University of Science and Technology has developed a novel 3D AMM-based tactile sensing platform that offers high-performance pressure sensing. The technology leverages auxetic metamaterials to enhance sensitivity, stability, and scalability.
Researchers developed a new origami-inspired folding strategy for reversible actuation of hydrogel pores, integrating facet-driven folding into polygonal pores to enable programmable and predictable actuation. This strategy retained 90% of its original shape after repeated swelling-shrinking cycles, demonstrating excellent reliability.
ConcreteSC technology achieves significant speed boosts and improved efficiency in next-generation wireless networks. The innovation integrates user tasks into communication processes, reducing computational complexity and increasing semantic meaning.
Researchers from Seoul National University of Science & Technology developed a smart adhesive system based on starfish for temporary and switchable underwater adhesion. The system exhibits high adhesion hysteresis, automatic release based on outside stimuli, and quick detachment by pneumatic actuation.
A study by Seoul National University researchers found that men and women prioritize different aspects of visibility in urban spaces, with men focusing on clear views and women concerned with unexpected threats. The research suggests designing piloti parking spaces with transparent fences and minimizing blind spots can reduce crime anx...
Seoul National University researchers create bioink from Kombucha SCOBY nanocellulose, suitable for in vivo tissue engineering. The bioink can be precisely applied directly onto damaged tissues using a digital biopen, paving the way for more personalized and effective wound healing.
A research team developed an autonomous geological assessment tool that uses machine learning to measure dip angles and directions in rock facets with high accuracy, achieving rates of up to 99.4%. The R-C-D-F method eliminates joint bands while preserving joint embedment points, making it ideal for modern infrastructure projects.
A Seoul National University team developed a breakthrough technique to improve metal-polymer joints, enabling the use of plastics in car parts and reducing overall vehicle weight. This method uses laser surface texturing to create strong and durable bonds between steel and plastic materials.
A new hybrid machine learning model predicts ultimate axial strength of CFRP-strengthened CFST columns with high accuracy, enabling safer and more efficient designs. The model can be used to optimize construction processes and enhance the safety of structures at a lower cost.
Researchers at Seoul National University of Science & Technology have made a breakthrough discovery in the catalytic recycling of polyolefins, enhancing conversion rates with the addition of water. The study reveals improved process efficiency, extended catalyst lifespan, and reduced operational costs.
The study proposes a novel approach for simultaneous extraction of metabolites and lipids from zebrafish embryos to shed light on PFOS-induced neurotoxicity. The researchers found that sphingolipids are a reliable biomarker of PFOS-induced neurotoxicity, and the method can be expanded to various biomolecules.
Researchers at SeoulNational University of Science & Technology propose two new designs for energy-efficient vibration energy harvesters that boost power output and efficiency. The designs use a repulsive magnet pair, yoke, and optimized coil placement to maximize magnetic flux change, leading to higher power generation.
Researchers at Seoul National University propose PV2DOC, a software that converts presentation videos into searchable, well-organized PDFs with text summaries and relevant images. The tool enhances video accessibility and reduces storage space, making it easier for users to find specific information.