446494 research outputs found
Sort by
Raisonnement diagnostique et connaissances : questions pratiques à poser lors de présentations de cas
No full textMessages-clés : La justesse diagnostique dépend davantage de l’accès à des connaissances pertinentes en mémoire que de l’utilisation de stratégies de résolution de problèmes. Ainsi, le clinicien-enseignant aura avantage à questionner les connaissances des étudiants à trois moments lors de présentations de cas, à savoir : lorsqu’ils explicitent leur première hypothèse diagnostique, lors de la sélection de signes cliniques discriminants à l’examen physique, et en fin de présentation concernant leurs incertitudes diagnostiques
Real-Time Multimodal Biometric Security with DIP-Based Preprocessing and Edge Deployment
No full textBiometric authentication is now crucial for secure digital access. However, unimodal methods still face issues like spoofing, background noise, and high cloud delays. Earlier multimodal models offered improved reliability but struggled with high computational demands and moderate accuracy levels (around 85-90%). This paper presents a new multimodal biometric authentication framework. It uses Deep Image Prior (DIP) for preprocessing, lightweight CNNs for feature extraction, and edge computing for real-time deployment. To boost performance, the model includes a Convolutional Block Attention Module (CBAM) and a Capsule Network layer. These components enhance the learning of unique features across fingerprint, iris, and facial types. The extracted features are combined using Fisher Vector with Gaussian Mixture Models (GMM) and classified through a quantized ResNet-101 backbone. Tests on the CASIA multimodal dataset, which includes over 10,000 samples, show that the proposed model reaches 96% accuracy. It surpasses current unimodal and multimodal systems and reduces latency by 40% on Raspberry Pi 4 and Jetson Nano platforms. The use of attentionguided and capsule layers sets this method apart from earlier models, providing a scalable solution that resists spoofing for banking, healthcare, defense, and IoT applications Index Terms—Biometric Authentication, Deep Image Prior, Multimodal Fusion, Edge Computing, Attention Mechanism, Capsule Netwo
Sustainable Boatbuilding: Evaluating the Feasibility of Recycled HDPE as a Structural Material
No full textThe boatbuilding industry is increasingly shifting toward environmentally sustainable materials. High-Density Polyethylene (HDPE) has become a highly viable alternative material because of its outstanding chemical and impact resistance, as well as its low-carbon manufacturing process. This makes it a practical substitute for traditional materials such as wood and fiberglass-reinforced plastic for constructing small vessels. However, the substantial generation of HDPE waste poses a significant challenge in maritime manufacturing. In response, this study presents a solution for the plastic waste produced during boatbuilding by recycling HDPE sourced from the Non-Metal Workshop at the Shipbuilding Institute of Polytechnic Surabaya (PPNS) and evaluating its suitability for use as a manhole cover. The compliance of recycled HDPE (r-HDPE) as a structural material was assessed based on the BKI (Biro Klasifikasi Indonesia) Standards for Thermoplastic Vessels, Volume 2, 2023 Edition. The mechanical testing of r-HDPE showed a tensile strength of 24.50 MPa and a flexural strength of 41.93 MPa. Finite Element Analysis using ANSYS showed that increasing the cover thickness from 6.5 mm to 12.5 mm significantly reduced deformation and stress, resulting in a safety factor of 8.29 under a maximum load of 1470 N. Therefore, 12.5 mm thick r-HDPE is structurally viable and supports sustainable shipbuilding by reducing waste
High-Resolution Coastal Topographic Model Using UAV Based Multi Georeferencing Method for Enhancing Sustainable and Resilient Coastal Zone Management
No full textCoastal zones are dynamic and vulnerable environments that require high-resolution topographic data to support sustainable and resilient management. This study aims to develop a high-resolution coastal topographic model using UAV photogrammetry with a multi-georeferencing approach, comparing Ground Control Points (GCP), Post-Processed Kinematic (PPK), and a combined GCP–PPK method. The research was conducted in the Kenjeran coastal area near the Suramadu Bridge, Surabaya, Indonesia using UAV survey. UAV surveys were performed to generate orthomosaics and digital terrain models (DTMs). Each georeferencing method was applied independently to evaluate its influence on spatial accuracy. Accuracy assessment used twelve Independent Check Points (ICPs), with horizontal and vertical accuracy evaluated through Circular Error (CE) and Linear Error (LE). The results indicate that the GCP-based model achieved the highest accuracy, with LE and CE values of 0.149 m and 0.264 m, respectively, outperforming both the PPK and combined methods. While PPK offers greater operational efficiency by reducing field control requirements, GCP-based georeferencing remains the most reliable for centimeter-level coastal mapping. The combined method provides a balanced trade-off between accuracy and efficiency. This study supports SDGs 13 (Climate Action) and 14 (Life Below Water) by enhancing climate-resilient coastal management
Feasibility Analysis of Decentralized Pressure Improvement in Peripheral Water Networks: Case Study of Zone 3, Surabaya, Indonesia
No full textThe study evaluates shifting Surabaya’s low-pressure coastal area (Zone 3) from a centralized to a decentralized water treatment system. Current residual pressure is below 2 mH2O, failing the 5–10 mH2O standard, and Scenario 1 shows the existing system cannot meet 2030 demand. Scenario 2, following the utility’s centralized expansion (Karang Pilang IV WTP, Putat Gede 3 Booster Pump, Mbah Ratu Reservoir), reaches 88% of customers at adequate pressure by 2030 but requires IDR 219.98 billion. A decentralized alternative with a local 500 L/s WTP and 900 m3 reservoir dedicated to Zone 3 achieves 98% coverage at the required pressure for only IDR 100.5 billion. The study concludes that decentralized systems can more effectively and cost‑efficiently boost pressure and meet growing demand where raw water sources are limited and centralized expansion is insufficient
Artificial Neural Network Backpropagation for Real-Time Coagulant Dose Prediction in Drinking Water Treatment
No full textA drinking water treatment plant is crucial for fulfilling the increasing water demand. As an integrated series, the coagulation unit is the most basic unit for removing particulates and reducing turbidity. However, the time gap to determine an effective coagulant dose was almost 6 h, which cannot accommodate the fluctuations in water inlet quality. A noteworthy method to reduce time is to use artificial neural network backpropagation (ANN-BP) to predict an optimum dose. The dataset consisted of five parameters (pH, temperature, conductivity, color, and turbidity) for a month of primary data sampling and historical jar test data from 2018 to 2022. The F-test result, F-value (6038,779) > F-table (2.21923), showed that one or more parameters had a statistically significant influence on the coagulant dose. Subsequently, a t-test excluded pH and temperature, with p-values lower than 0.05. Empirical models were developed through trial-and-error variations of the input layers (three and five parameters), hidden layers (2-10 nodes), and an output. The models with the lowest MSE and highest R2 were [5-6-1] (R2 =- 0.96051; MSE = 0.00179) and [3-4-1] (R2 = 0.97755; MSE =0.00102). In conclusion, [3-4-1] is recommended because it has the lowest MSE and the highest R2
Early Afterslip Deformation Captured by Sub-Daily Kinematic GPS following the 2010 Mw 7.8 Mentawai Earthquake
No full textOn 25 October 2010, an Mw 7.8 earthquake occurred west of the Mentawai Islands, releasing energy from a shallow offshore fault and generating complex ground deformation patterns. Although surface deformation has been monitored for over a decade, the precise geometry of the slipping zone remains unresolved. This study investigates the event using sub-daily kinematic GPS observations from the Sumatra GPS Array (SuGAr) to analyze coseismic and early postseismic deformation following the rupture. The resulting displacement field reveals limited horizontal motion of approximately 20–25 cm and vertical subsidence of about 4–6 cm near the islands, suggesting that most slip was concentrated offshore toward the trench. In the subsequent days, the GPS time series displays a gradual decay in motion, which can be modeled with a logarithmic function and characteristic timescales of roughly five days. This transient response is interpreted as an early afterslip evolving downdip of the main rupture area. These findings underscore the heterogeneous nature of the plate interface, where seismic and aseismic processes occur concurrently over short timescales. Future work will estimate early afterslip using a time-dependent inversion model to further characterize the rupturing zone and clarify the transition between seismic and aseismic deformation
Big Data-Based Literature Study on Automatic Identifition System Data and Synthetic Aparture Radar Image Integration for Illegal Fishing in Maritime Awareness
No full textIllegal, unreported, and unregulated (IUU) fishing remains a persistent threat in Indonesian waters, causing substantial economic losses and long-term ecological damage. This review synthesizes methods for fusing Automatic Identification System (AIS) data with Synthetic Aperture Radar (SAR) imagery to enhance maritime surveillance. AIS conveys vessel identity and reported position, whereas SAR detects vessels operating without AIS (“dark” vessels). The review covers approaches to spatiotemporal synchronization, data association, and machine-learning models that jointly exploit both modalities. In addition, this study provides a systematic mapping of recent AIS–SAR fusion methods and proposes a conceptual big data framework tailored to Indonesia’s maritime surveillance context. According to the surveyed literature, AIS–SAR fusion has been reported to improve the identification of non-cooperative vessels, reduce false alarm and missed detection rates, and shorten response times. Effective implementation requires reliable spatiotemporal alignment, adequate computing resources for large-scale processing, and interagency data-sharing mechanisms. Collectively, the evidence indicates that large-scale AIS–SAR fusion can enhance maritime awareness and support Indonesia’s efforts to counter IUU fishing
Analysis of Residential Land Changes Using Object Based Image Analysis (OBIA) Method with SPOT Imagery in 2014 and 2024 (Case Study: Banyumanik, Semarang City)
No full textResidential land expansion in urban areas, particularly in Banyumanik District, Semarang City, has increased significantly over the past decade. This study analyzes residential land-use changes between 2014 and 2024 using Object-Based Image Analysis (OBIA) applied to SPOT 7 satellite imagery and examines the influence of public and social facilities on these changes. An integrated spatial and statistical approach was employed. Residential land-use maps for 2014 and 2024 were generated and overlaid with public and social facility maps to assess spatial relationships, followed by statistical analysis to evaluate their influence on residential expansion.The results indicate a substantial increase in residential land area, predominantly in locations with high accessibility to public and social facilities. Statistical findings confirm that facility availability significantly influences residential land expansion, highlighting infrastructure provision as a key driver of urban growth. These findings imply that urban development can be more effectively managed by aligning the spatial distribution of public and social facilities with land-use planning and zoning regulations. Such integration is essential to promote sustainable urban growth and to mitigate uncontrolled residential expansion in Semarang City
Fractionation of cellulose and lignin from sugarcane bagasse via the alkaline and acid chemical process
No full textThis research aimed to fractionate cellulose and lignin from sugarcane bagasse (SCB) using the alkaline and acid chemical process. Sulphuric acid (H2SO4) and Sodium hydroxide (NaOH) were selected as the pretreatment chemical solvents. After pretreatment, the cellulose content and lignin were 98.5% and 79.2%, respectively. Fourier Transform Infrared (FTIR) Spectroscopy, 1H and 13C Nuclear Magnetic Resonance (NMR), and Scanning Electron Microscopy (SEM) were employed to analyse the material. Raw SCB shows the vibration absorption peaks of O-H and C-H stretching at 338.23 and 2897.37 cm-1, which are correlated with cellulose, hemicellulose, and lignin molecules. The vibration of O-H and C-H functional groups was observed on extracted cellulose and lignin biopolymers. However, the intensity of vibration peaks decreased, especially for lignin biopolymer. The absorption band observed at 1603.74 and 1541 cm-1 was attributed to the aromatic (C=O, C=C) structure. 1H and 13C NMR spectra detected the presence of both the aromatic regions and the sidechain regions on raw SCB and SCB-lignin. However, only the sidechain regions were detected on SCB-cellulose. SEM analysis reveals that the surfaces of raw SCB, SCB-cellulose, and SCB-lignin exhibit smooth, irregular, and uneven shapes, respectively. The results show that cellulose and lignin were simultaneously recovered from SCB