446494 research outputs found
Sort by
A robust SAR change detection pipeline for landslide mapping using tile-wise HFEM and graph-cut Refinement
No full textLandslide detection from SAR images remains challenging due to speckle noise, varied terrain conditions, and limited annotated data. In this work, a hybrid model with a Histogram Feature Extraction Module (HFEM) and Graph-Cut Refinement is used to improve change detection. The dataset consists of pre- and post-event SAR images of the 2021 Haiti earthquake with ground-truth annotations. The HFEM is used to acquire discriminative spatial features and suppress noise, and Graph-Cut Refinement is used to regularize segmentation and make it more consistent. The experimental results verify that the proposed approach obtains high recall and comparative precision, balanced F1-score, and fewer error regions than baseline approaches. The qualitative findings also demonstrate the steadiness of the forecast change maps. Overall, the system is very suitable for geospatial mapping of big landslides and supports disaster relief and risk management uses
Thermal Performance Evaluation of Bio-Based Insulation Materials in Residential Buildings Across Continental Climates
No full textThis study the thermal and economic analysis of four bio-based insulations, including hempcrete, sheep wool, cellulose fiber, and straw bale, in residential constructions in three regions in continental climates: Uzbekistan, Kazakhstan, and Mongolia. An open-source Python based simulation framework was created to reduce the overall cost, which entails the life-cycle energy cost and material installation cost in 15 years. Analysis was incorporated with climate-specific parameters, including Heating and Cooling Degree Days (HDD and CDD) or electricity price. Findings indicate that cellulose fiber has recorded the lowest cost for a total of the three regions with the optimum thickness of 0.21 m in Uzbekistan, and 0.35 m in Mongolia. Hempcrete had the least thermal conductivity and material density; hence, it was the most expensive with the highest thickness. The paper shows that the cost-effectiveness of the energy retrofits can be greatly increased with the help of climate-adapted insulation design developed with the help of simulation-based optimization. The results can be translated into practical recommendations to policy-makers and architects interested in the creation of low-carbon and thermally efficient solutions in residential construction in a variety of climate zones
A MediaPipe and CNN-Based Framework for Real-Time Sign Language Recognition and Translation
No full textThis research is being aimed at the real-time sign language translation system developed in the absence of dedicated equipment like sensor gloves, under the sole observation of a standard camera and computer vision techniques. The project was divided into two major phases: Static Sign Recognition and Dynamic Sign Recognition. During the static phase, a fully original dataset comprising 120,000 images was created for 24 alphabets (A–Z excluding J and Z). Each image was then treated through two custom-designed filters: silhouette extraction and finger detection, carried out using MediaPipe for precise annotation and cropping on its own. The training of the static model happened with custom-designed CNN applied on the data to achieve accuracies of 94% (finger detection), 89% (silhouette), and 91% (combined). In the dynamic phase, signs with movement were captured using MediaPipe landmarks and registered as NumPy arrays (CSV and .NPY) files. Hand and facial keypoints were stored at every frame for temporal learning. This phase included one-hand, two-hand, and hand-with-face categories. The whole system achieved 15- 20 FPS real-time inference on CPU alone. While the static signs and single-handed dynamic signs were performing very well, the accuracy of complex two-handed and face-assisted gestures suffered due to scarcity and noise in the data. The results provide a very good baseline to improve upon for the future using deep temporal networks like LSTM or 3D CNNs
Diabetes Prediction using Deep Learning: An Analysis of the Pima Indian Dataset
No full textDiabetes mellitus is a significant global health concern, and its early and correct diagnosis is required to prevent significant health-related complications in the long run. It is composed of the Pima Indians Diabetes Dataset, which is used to deepen the understanding of diabetes prediction with the help of the deep learning algorithm. There is a lot of preprocessing, including the feature selection through the Extra Trees Classifier, the normalization, the treatment of the missing records or the zero values, and the elimination of the duplicate records. It uses three deep learning models as base learners, namely Convolutional Neural Network (CNN), Long Short-term Memory (LSTM), and Artificial Neural Network (ANN). Key metrics are used to estimate their performance in the aspect of predictive strength. In this strategy, an Ensemble Deep Learning method aims at increasing the credibility of such models by layering their results on top of a meta-level classifier. Experimental evidence demonstrates that the stacked ensemble is better than individual models in every instance and exploits the complementary advantages of the latter. The work demonstrates how ensemble deep-learning may facilitate the creation of powerful clinical decision-support systems to make early diagnosis and prevention of diabetes
Design And Analysis Of 5 Stage Pipelined CPU With Hazard Handling
No full textThis paper compares two different RISC instruction set architectures, one with minimal hazard control and the other with enhanced design, using forward, interlock and special register hazard handling techniques to minimize the effects of hazards on the operating speed of both systems. Both systems were modelled and synthesized using the Xilinx Vivado Toolchain. The results showed a 92.4% decrease in the time required to resolve a hazard, as well as a 38% decrease in stall cycles needed in the enhanced processor compared to the basic processor. The final synthesis of both designs placed them on a Xilinx Artix-7 Field Programmable Gate Array (FPGA) and determined the maximum operating frequency of 167 MHz for the enhanced design, using approximately 12,800 look-up tables (LUTs). Based on this information, the authors conclude that the enhancements to the processor will increase performance and user-friendliness for embedded applications requiring high performance
Multi-User Real Time Polling System: Enhancing Online Decision-Making With MERN Stack
No full textThe digital polling and e-voting systems are the main reasons that have led to the development of this type of systems. We have presented an all-inclusive real-time voting platform based on the MERN stack, which consists of MongoDB, Express.js, React.js, and Node.js.This project has able to handle massive traffic with excellent responsiveness. The users can effortlessly and graphically view the real-time results through charts, polls, and managing different polls. AI-based opinion mining is another tool that helps the platform in delivering more accurate insights. It keeps on playing a significant role in the sentiment classification as well as refining the trends based on the feedback given. Blockchain technology is an additional factor that assures security and trustworthiness regarding the certified votes and even if it is later embraced, it can turn out to be an extremely powerful security technique. The real-time updates are carried out via Web Socket communication, which ensures that all users have the latest counts and results at their disposal. Community polling is one more attribute of this platform which simplifies the process of getting the most accurate feedback among schools, business groups, and other user-based communities. This platform is not only a fortress but also an ingenious way of interacting with users, and it not only gets rid of the limitations of existing e-voting systems but also offers constructive feedback that aids in the decision-making process by making it more efficient
Clean tracks: Automated hygiene system for Indian railway restrooms
No full textThe hygiene of toilets in railways presents an ongoing challenge. This paper introduces Clean Tracks, a fully automated sanitization system using ultraviolet sterilization along with hydrogen sulphide (H2S) sensors and smoke/alcohol sensors. Automated flushing and sensor monitoring with proactive maintenance notifications is facilitated in real-time, resulting in more proactive management of toilet hygiene. The findings show that the toilet system was effective in reducing odor, microbial burdens, and increasing user satisfaction and offers a new scalable, smart sanitation option for railways
The influence of bidirectional tidal flow on scour mechanisms and morphological changes around the Suramadu bridge piers
No full textThe Suramadu Bridge is a vital infrastructure in the Madura Strait, facilitating regional economic connectivity. However, complex oceanographic characteristics—notably bidirectional currents and seasonal variations—pose significant scouring risks to bridge piers, potentially threatening structural integrity. This study models current patterns and sediment transport to analyze scouring dynamics around the piers. Using hydrodynamics model, simulations were conducted for 2024, incorporating bathymetry, tides, wind, and sediment data. Model validation against tidal data showed high accuracy (MAE: 0.1473 m; RMSE: 0.1771 m). Results reveal a tidal-driven west–east bidirectional flow with stable average velocities (0.21–0.25 m/s) across monsoons. Interaction between currents and piers causes upstream flow deceleration and lateral acceleration, triggering deposition and erosion, respectively. Piers P46, P47, and P49 experienced the most significant scouring, with maximum depths reaching -0.250 m, while P37, P41, and P56 remained stable. Erosion trends were spatially and temporally consistent across both west and east monsoons. These findings establish critical monitoring priorities to ensure the bridge’s structural sustainability in alignment with SDG Goal 9
Determinants of Enhanced Trained Human Resources in Flood Disaster Risk Reduction (A Case Study in Sidomulyo Village, Lamongan)
No full textIn many developing regions, top-down infrastructural solutions fail to mitigate recurrent disasters, such as the annual Bengawan Jero flood in East Java, Indonesia, which causes estimated losses of IDR 29 billion (approximately USD 1.8 million). This failure underscores the critical need for community-level capacity, yet Sidomulyo Village remains a “cold spot” with a severe deficit in trained human resources. This research aims to identify the key determinants for enhancing these resources. Employing Partial Least Squares-Structural Equation Modelling (PLS-SEM), the roles of local institutions, social capital and human capital were assessed. The analysis shows that human capital has a strong, direct positive effect on trained human resources (β = 0.644, p value 0.000), while local institutions and social capital have no significant direct effect. However, the study uncovers a critical indirect path wherein social capital significantly impacts trained human resources by mediating the development of human capital (β = 0.343, p value 0.013). The clear conclusion is that effective strategies for enhancing trained human resources must focus on building human capital while using social capital as the fundamental supporting mechanism
The Impact of English Teachers’ Classroom Management Style on Self-Control among Primary School Students
No full textAs China keeps carrying out the “Double Reduction” Policy, the importance of quality and efficiency is increasingly highlighted in primary school. In English classes, with the need for students’ continuous input and timely feedback, there is a rising demand for self-control ability. Self-control ability, as an important part of students’ overall development, is performed to different extents in different teachers’ classroom management styles, including authoritarian, authoritative, and permissive. Therefore, exploring the relationship between English teachers’ classroom management styles and students’ self-control has become one of the key issues in primary school. This research focuses on the domain of the primary English classroom in China, complementing the context of the school, especially classes, and exploring the characteristics of different classroom management styles and differences in the impact of these styles on self-control behaviors, which also provides a reference for balancing management and education in future education. This study employs various research methods, such as literature review and mechanism description. Furthermore, comprehensive strategies and methods will be discussed. Corresponding conclusions are drawn based on the research findings