Development Of Real-time System Research Articles

IMPLEMENTASI BIG DATA ANALYTICS DALAM KLASIFIKASI KUALITAS UDARA MENGGUNAKAN ALGORITMA GRADIENT-BOOSTED TREE CLASSIFIER PADA PYSPARK

This study aims to classify air quality based on PM1.0, PM2.5, and PM10 parameters using a Big Data Analytics approach with the Gradient-Boosted Tree Classifier (GBT) algorithm implemented on the PySpark framework. The dataset used was downloaded from OpenAQ, covering the period from April 14, 2021, to April 16, 2023, with a total of 1,048,154 entries, representing a large and complex volume of data. The research process includes data preprocessing to address data imbalance, dataset splitting for training and testing, and hyperparameter tuning using grid search and cross-validation to optimize model performance. By leveraging PySpark’s advantage in parallel processing of large data, the GBT model achieved an accuracy of 98.87%, precision of 99.00%, recall of 98.87%, and an F1-Score of 98.90%. This study demonstrates how Big Data Analytics can enhance efficiency and accuracy in air quality classification, contributing significantly to the development of real-time monitoring systems that support air pollution mitigation and data-driven policy-making.

Development and testing of a dual-frequency real-time hardware feedback system for the hard X-ray nanoprobe beamline of the SSRF.

A novel dual-frequency real-time feedback system has been developed to simultaneously optimize and stabilize beam position and energy at the hard X-ray nanoprobe beamline of the Shanghai Synchrotron Radiation Facility. A user-selected cut-off frequency is used to separate the beam position signal obtained from an X-ray beam position monitor into two parts, i.e. high-frequency and low-frequency components. They can be real-time corrected and optimized by two different optical components, one chromatic and the other achromatic, of very different inertial mass, such as Bragg monochromator dispersive elements and a pre-focusing total external reflection mirror. The experimental results shown in this article demonstrate a significant improvement in position and energy stabilities. The long-term beam angular stability clearly improved from 2.21 to 0.92 µrad RMS in the horizontal direction and from 0.72to 0.10 µrad RMS in the vertical direction.

Spectroscopy-Based Methods and Supervised Machine Learning Applications for Milk Chemical Analysis in Dairy Ruminants

Milk analysis is critical to determine its intrinsic quality, as well as its nutritional and economic value. Currently, the advancements and utilization of spectroscopy-based techniques combined with machine learning algorithms have made the development of analytical tools and real-time monitoring and prediction systems in the dairy ruminant sector feasible. The objectives of the current review were (i) to describe the most widely applied spectroscopy-based and supervised machine learning methods utilized for the evaluation of milk components, origin, technological properties, adulterants, and drug residues, (ii) to present and compare the performance and adaptability of these methods and their most efficient combinations, providing insights into the strengths, weaknesses, opportunities, and challenges of the most promising ones regarding the capacity to be applied in milk quality monitoring systems both at the point-of-care and beyond, and (iii) to discuss their applicability and future perspectives for the integration of these methods in milk data analysis and decision support systems across the milk value-chain.

Development of a comprehensive air risk index and its application to high spatial-temporal health risk assessment in a large industrial city.

Particulate matter (PM) contains various hazardous air pollutants (HAPs) that can adversely affect human health, highlighting the need for an integrated index to represent the associated health risks. In response, this study developed a novel index, the comprehensive air-risk index (CARI), for Ulsan, the largest industrial city in South Korea. This index integrates toxicity-weighted concentrations of polycyclic aromatic hydrocarbons (PAHs) and heavy metals using their inhalation unit risks. CARI was categorized into four risk levels based on probabilistic health risks. Over eight years (2013-2020) in Ulsan, the risk from PAH exposure showed a decreasing trend, whereas the risk from heavy metals remained stable, reflecting different emission patterns and major source types. PAHs and heavy metals contributed 38.1% and 61.9% to CARI, respectively, highlighting the greater impact of heavy metals on human health. Unlike the monthly variations in PM2.5 concentrations, CARI values tended to increase in the summer and decrease in the spring and fall, indicating the impact of local emissions, particularly from petrochemical and non-ferrous industrial facilities. Moreover, a machine learning model enhanced the spatio-temporal resolution of CARI, showing that 'unhealthy' days were 2.4 times more frequent in industrial areas than in urban areas. In conclusion, CARI is a promising tool for assessing health risks in industrial cities and for developing risk-based management plans. Furthermore, we propose the development of a national-scale real-time CARI system by enhancing the spatio-temporal resolution of HAP data through the use of machine learning.

High-performance single SiC nanocable-based plasmonic photodetectors for ultraviolet communication systems

Silicon carbide (SiC) nanowires are the most promising candidate for developing high-performance single-nanowire-based UV photodetector (PD) with outstanding photoelectronic properties and low power consumption. However, SiC single-nanowire-based UV PDs commonly suffer from very low light current on the order of pA or nA, requiring precision equipment or an extra current amplification circuit, which greatly limit their practical application. To overcome this bottleneck, novel SiC single nanocable-based plasmonic PDs with the light current on the order of mA were successfully developed with the features of improving light absorption and reducing the dark current. The SiC/SiO2@Ag nanocables consisting of SiC nanowire core and uniform SiO2 shell encrusted with scattered and isolated Ag nanoparticles (NPs) were fabricated by a simple, low-cost, and space-confined vacuum-heating strategy using ultralong SiC nanowires and silver nitrate precursor followed by the annealing process. In such architecture, the in situ resulting SiO2 shell can effectively passivate the surface defects of the SiC nanowire core, restrain the photo-excited carrier's losses, and effectively block the hot electron injection, leading to a great reduction in the dark current and enhancement in the light current. The encrusted Ag NPs on the nanocable surface exhibited a strong LSPR effect with significantly increased optical absorption. Benefiting from the synergistic effect of SiO2 passivation and Ag nanoparticle LSPR effect, the device current increased dramatically by several orders of magnitude to reach 0.37 mA at a bias voltage of 3 V. Moreover, the developed SiC/SiO2@Ag PDs had faster response speed (0.27 s), lower dark current at the nA level, and high stability, which can be competent for the development of real-time, accurate, and cost-effective communication systems and flame detection with impressive switching ratio as high as 66012.

Self-Rectifying Memristor-Based Reservoir Computing for Real-Time Intrusion Detection in Cybersecurity.

The increasing sophistication of cybersecurity threats, driven by the proliferation of big data and the Internet of Things (IoT), necessitates the development of advanced real-time intrusion detection systems (IDSs). In this study, we present a novel approach that integrates NiO-doped WO3-x/ZnO bilayer self-rectifying memristors (SRMs) within a reservoir computing (RC) framework for IDS applications. The proposed crossbar array architecture exploits the exceptional dynamic properties of SRMs, achieving a classification accuracy of 93.07% on the CSE-CIC-IDS2018 data set, while demonstrating ultrahigh information-processing efficiency. Our approach not only leverages the tunable characteristics of memristors but also addresses the challenge of sneak path currents in large-scale integration, offering a robust and scalable solution for next-generation IDS. This work exemplifies the power of emerging electronics in enhancing cybersecurity through innovative hardware implementations.

Machine vision combined with deep learning-based approaches for food authentication: An integrative review and new insights.

Food fraud undermines consumer trust, creates economic risk, and jeopardizes human health. Therefore, it is essential to develop efficient technologies for rapid and reliable analysis of food quality and safety for food authentication. Machine vision-based methods have emerged as promising solutions for the rapid and nondestructive analysis of food authenticity and quality. The Industry 4.0 revolution has introduced new trends in this field, including the use of deep learning (DL), a subset of artificial intelligence, which demonstrates robust performance and generalization capabilities, effectively extracting features, and processing extensive data. This paper reviews recent advances in machine vision and various DL-based algorithms for food authentication, including DL and lightweight DL, used for food authenticity analysis such as adulteration identification, variety identification, freshness detection, and food quality identification by combining them with a machine vision system or with smartphones and portable devices. This review explores the limitations of machine vision and the challenges of DL, which include overfitting, interpretability, accessibility, data privacy, algorithmic bias, and design and deployment of lightweight DLs, and miniaturization of sensing devices. Finally, future developments and trends in this field are discussed, including the development of real-time detection systems that incorporate a combination of machine vision and DL methods and the expansion of databases. Overall, the combination of vision-based techniques and DL is expected to enable faster, more affordable, and more accurate food authentication methods.

Leather-Based Shoe Soles for Real-Time Gait Recognition and Automatic Remote Assistance Using Machine Learning.

Real-time monitoring of gait characteristics is crucial for applications in health monitoring, patient rehabilitation feedback, and telemedicine. However, the effective and stable acquisition and automatic analysis of gait information remain significant challenges. In this study, we present a flexible sensor based on a carbon nanotube/graphene composite conductive leather (CGL), which uses collagen fiber with a three-dimensional network structure as the flexible substrate. The CGL-based sensor demonstrates a high dynamic range, with notable pressure responses ranging from 0.6 to 14.5 kPa and high sensitivity (S = 0.2465 kPa-1). We further developed a device incorporating the CGL-based sensor to collect foot characteristic signals from human motion and designed smart sports shoes to facilitate effective human-computer interaction. Machine learning was employed to collect and process gait characteristic information in various states, including standing, sitting, walking, and falling. For real-time monitoring of falls, we optimized the K-Nearest Time Series Classifier (KNTC) algorithm, achieving an accuracy of 0.99 and a prediction time of only 13 ms, which highlights the system's excellent intelligent response capabilities. The system maintained a gait recognition accuracy of 90% across diverse populations, with low false-positive (3.3%) and false-negative (3.3%) rates. This work demonstrates stable gait recognition capabilities and provides valuable methods and insights for plantar behavior monitoring and data analysis, contributing to the development of advanced real-time gait monitoring systems.

Intraoperative navigation in craniofacial surgery.

Craniofacial surgery requires comprehensive anatomical knowledge of the head and neck to ensure patient safety and surgical precision. Over recent decades, there have been significant advancements in imaging techniques and the development of real-time surgical navigation systems. Intraoperative navigation technology aligns surgical instruments with imaging-derived information on patient anatomy, enabling surgeons to closely follow preoperative plans. This system functions as a radiologic map, improving the accuracy of instrument placement and minimizing surgical complications. The introduction of first-generation navigation systems in the early 1990s revolutionized surgical procedures by enabling real-time tracking of instruments using preoperative imaging. Initially utilized in neurosurgery, intraoperative navigation has since become standard practice in otolaryngology, cranio-maxillofacial surgery, and orthopedics. Since the 2000s, second-generation navigation systems have been developed to meet the growing demand for precision across various surgical specialties. The adoption of these systems in craniofacial surgery has been slower, but their use is increasing, particularly in procedures such as foreign body removal, facial bone fracture reconstruction, tumor resection, and craniofacial reconstruction and implantation. In Korea, insurance coverage for navigation in craniofacial surgery began in 2021, and new medical technologies for orbital wall fracture treatment were approved in August 2022. These technologies have only recently become clinically available, but are expected to play an increasingly important role in craniofacial surgery. Intraoperative navigation enhances operative insight, improves target localization, and increases surgical safety. Although these systems have a steep learning curve and initially prolong surgery, efficiency improves with experience. Calibration issues, registration errors, and soft tissue deformation can introduce inaccuracies. Nonetheless, navigation technology is evolving, and the integration of intraoperative computed tomography data holds promise for further enhancements of surgical accuracy. This paper discusses the various types and applications of navigation employed in craniofacial surgery, highlighting their benefits and limitations.

A real-time eye movement-based computer interface for people with disabilities

It is costly to develop systems that enable individuals exposed to Amyotrophic Lateral Sclerosis and similar diseases that directly affect the neuromotor ability to communicate with the outside world. In this study, a budget friendly, high-accuracy, software-based, gaze-controlled, real-time virtual keyboard approach that can enable these people to communicate effectively is proposed. The proposed application requires only a computer and a webcam and has a user-friendly interface that meets the basic daily needs of individuals with disabilities. Since the proposed system does not require an extra action such as blinking, it makes it possible to use computers in advanced stage patients who cannot blink their eyes. The application which uses a deep learning-based facial landmark detector, determines the letters the user focuses on the screen and converts thoughts into text. The part of the screen that the user focuses on is determined with a new selection approach inspired by the K-Nearest Neighbors algorithm. This approach, which does not require blinking, offers high speed and accuracy. In the tests, a typing speed of 23.33 characters per minute is achieved with an accuracy rate of 95.12%. It is anticipated that the study will increase computer accessibility for disabled individuals with limited mobility and contribute to the development of real-time eye tracking systems.

Improving preparation in the emergency trauma room: the development and impact of real-time data transfer and dashboard visualization system.

This study examines, with clinical end users, the features of a visualization system in transmitting real-time patient data from the ambulance to the emergency trauma room (ETR) to determine if the real-time data provides the basis for more informed and timely interventions in the ETR before and after patient arrival. We conducted a qualitative in-depth interview study with 32 physicians in six German and Swiss hospitals. A visualization system was developed as prototype to display the transfer of patient data, and it serves as a basis for evaluation by the participating physicians. The prototype demonstrated the potential benefits of improving workflow within the ETR by providing critical patient information in real-time. Physicians highlighted the importance of features such as the ABCDE scheme and vital signs that directly impact patient care. Configurable and mobile versions of the prototype were suggested to meet the specific needs of each clinic or specialist, allowing for the transfer of only essential information. The results highlight on the one hand the potential need for adaptable interfaces in medical communication technologies that balance efficiency with minimizing additional workload for emergency medical services and show that the use of pre-notification systems in communication between ambulance and hospital can be supportive. Further research is recommended to assess practical application and support in clinical practice, including a re-evaluation of the enhanced prototype by professionals.

Applied static analysis and specialization of cross-core syscalls for multi-core AUTOSAR OS

The development of static real-time control systems often follows a closed-world assumption, allowing extensive RTOS-aware whole-program optimization. For single-core systems, previous work could show the high potential of control-flow-aware static system-call tailoring. However, due to an exponential state explosion in the analysis phase, it cannot simply be extended to a multi-core setting, since the core’s relative timing to each other is undetermined. In this work, we present MultiSSE, a multi-core capable and RTOS-aware static whole-system optimization. First, MultiSSE analyzes the system by determining the relative positions of multiple cores only when necessary. For that, it exploits structural control flow and optionally timing information to handle each core separately as much as possible. Based on the analysis result, a synthesis applies lock elision and system-call optimization to generate specialized multi-core real-time systems for AUTOSAR OS. To enable a static prediction of the run-time reduction, we additionally provide cost models for the optimized cross-core system calls and evaluate the approach with synthetic benchmarks and a real-world quadrotor application. MultiSSE was able to optimize or even completely elide costly cross-core system calls and system objects leading to a reduction of up to 14% of a task’s execution time. In this extended version of a conference publication (Entrup et al. 2023), we provide an advanced description, new cost models, and an end-to-end measurement by developing a synthesis complementing the analysis.

Real-time Vertical Air Quality Monitoring System Development for Urban Scale

Real-time Vertical Air Quality Monitoring System Development for Urban Scale

Development of real-time energy monitoring system using IoT base

This paper aims to develop a real-time energy monitoring system based on smart metering to enhance energy efficiency in the single-phase residential sector. Based on the concept of low-cost IoT devices, this intelligent meter system is designed to monitor household energy consumption. It provides real-time information on a graphical Node-RED Dashboard, making it easy for households to track and manage their energy usage. The hardware of the power metering system included the Node MCU ESP8266, the PZEM-004T, and a cloud server built on the Raspberry Pi for storing electricity consumption data by using Node-RED to connect devices via an application programming interface system. This system can help analyze the electricity consumption behavior in the residential sector. It is a guideline for selecting the electricity rate between the TOD and TOU rates. The results found that in the electrical energy measurements of households 1 and 2, the mean deviations were 0.8758% and 0.5523%, respectively. The electricity cost-saving results when changing the electricity tariff from the TOD rate of households 1 and 2 to the TOU rate. It was found that electricity costs can be saved by 0.2807% and 1.0936%, respectively. The most critical variable is electricity consumption during peak periods. In the case study of household 1, if the electricity consumption during the peak period is less than 13%, it will be appropriate to select the Time of Use rate. In household 2, if there is electricity consumption during the peak period, less than 38% would be appropriate to choose the type of electricity user with a TOU rate.

Development and application of multi-parameter real-time acquisition system for intelligent secondary equipment

Abstract A multi-parameter real-time collection and analysis system, including intelligent secondary equipment, data collection gateway, and health management backend, has been constructed to solve the problem of current incomplete monitoring data and inability to providential warning of hidden faults of intelligent secondary equipment in substations. The key feature parameters of the device’s functional modules were extracted, and a data collection gateway was developed to achieve IoT access to feature data. A fault identification and warning model for the secondary equipment functional module considering temperature is provided, which realizes the assessment of equipment health status and ensures the stable and reliable operation of the power grid.

Advanced Techniques for Real-time Facial Expression Recognition Using Deep Learning

Abstract—Developing systems that can automatically recognize and interpret human emotions from facial expressions is the aim of the quickly expanding field of facial emotion identification and detection research. This technology finds applications in a wide range of areas, including as healthcare, marketing, security, and human-computer interface. Using computer vision and machine learning algorithms, facial emotion recognition systems analyze a face’s features and classify it into numerous emotional categories, including joyful, sorrowful, angry, fearful, and surprised.The three steps in the multi-step process that goes into identifying facial emotions are face detection, facial feature extraction, and emotion categorization. Thanks to recent advances in deep learn- ing, facial emotion detection systems can now identify emotions with high resilience and precision. Further, the development of real-time face expression recognition systems has opened up new avenues for applications such as sentiment analysis, emotional intelligence, and affective computing. This technology could fundamentally alter human-machine interactions and open the way to more compassionate and personalized relationships.A multitude of applications, including virtual assistants, mental health aids, and human-centered technology, will be greatly impacted by the development of systems for identifying and de- tecting facial expressions. Artificial intelligence (AI) technologies that recognize emotions allow people to interact with the digital world in more intelligent and flexible ways. But the complexity of emotion identification lies in the fact that it requires context and geometric elements in addition to facial expressions.

Real-Time Detection of Face Mask Usage Using Convolutional Neural Networks

The widespread adoption of face masks has been a crucial strategy in mitigating the spread of infectious diseases, particularly in communal settings. However, ensuring compliance with mask-wearing directives remains a significant challenge due to inconsistencies in usage and the difficulty in monitoring adherence in real time. This paper addresses these challenges by leveraging advanced deep learning techniques within computer vision to develop a real-time mask detection system. We have designed a sophisticated convolutional neural network (CNN) model, trained on a diverse and comprehensive dataset that includes various environmental conditions and mask-wearing behaviors. Our model demonstrates a high degree of accuracy in detecting proper mask usage, thereby significantly enhancing the ability of organizations and public health authorities to enforce mask-wearing rules effectively. The key contributions of this research include the development of a robust real-time monitoring system that can be integrated into existing surveillance infrastructures to improve public health safety measures during ongoing and future health crises. Furthermore, this study lays the groundwork for future advancements in automated compliance monitoring systems, extending their applicability to other areas of public health and safety.

Applicability of pre-trained CNNs in temperate deforestation detection

ABSTRACT Technological advancements have opened up new possibilities for accurately identifying deforested areas and developing effective data collection strategies. Our paper proposes a practical approach to improve deforestation monitoring using satellite images and Convolutional Neural Networks (CNNs) for image classification. To train the CNNs, we used labeled data primarily sourced from the Amazon basin, while images from the Romanian Carpathian mountains were reserved for testing. Building on this foundation, we evaluated various pre-trained CNN architectures, including AlexNet, VGGNet, ResNet, DenseNet, EfficientNet, GoogLeNet, and Swin Transformer, comparing their performance with a typical CNN architecture. In addition, we refined the performance by testing four ensemble learning methods. We found a decrease of only 10% in accuracy for using the models on data from the temperate area. Our findings contribute to this interdisciplinary field by providing insights into the effectiveness of pre-trained CNNs in classifying satellite images from temperate regions, trained with knowledge from tropical deforestation data. This contribution may open ways for the development of a quasi real-time deforestation monitoring system and serve as a reference for future research in temperate areas.

The Electrostatic Induction Characteristics of SiC/SiC Particles in Aero-Engine Exhaust Gases: A Simulated Experiment and Analysis

This study investigates the electrostatic induction characteristics of silicon carbide-fiber-reinforced silicon carbide (SiC/SiC) particles within aero-engine exhaust gases using a dedicated J20 turbojet engine experimental platform. Our comprehensive experiments explored the electrostatic properties of SiC/SiC particles under varying engine operational states—specifically focusing on different thermal conditions, particle mass concentrations, particle sizes, and exhaust gas velocities compared to those of common engine exhaust constituents like carbon (C) and iron (Fe) particles. The results demonstrate that SiC/SiC particles consistently maintain a stable positive charge across varied temperatures, significantly diverging from the behaviors of carbon (C) and iron (Fe) particles. Additionally, our findings reveal that higher mass concentrations of SiC/SiC particles, smaller particle sizes within a certain range, and greater exhaust gas velocities of the aero-engine all lead to increased particle charge and more pronounced electrostatic induction characteristics. This study highlights the potential of electrostatic sensors for the early detection and diagnosis of failures in aero-engines, offering crucial insights into the development of more resilient real-time aero-engine health monitoring systems.

Maximized nanojunctions in Pd/SnO2 nanoparticles for ultrasensitive and rapid H2 detection

H2 energy has gained massive attraction as a promising candidate for future energy sources. However, the explosive properties of H2 arouse significant safety concerns, emphasizing the need for the development of real-time H2 detection systems. This study presents the fabrication of an ultrasensitive and selective H2 gas sensor using Pd/SnO2 nanoparticles (NPs) synthesized through the utilization of Pluronic F-127. Pluronic F-127 improves the dispersion and regulates the particle size of Pd NPs. Pd/SnO2 NPs, which are comprised of numerous nanojunctions, exhibit H2 response of 27,190 and a response time of 3 s when exposed to 50 ppm of H2 at 100 °C. The practical applicability of Pd/SnO2 NPs was demonstrated by detecting H2 generated from water-splitting cells, exhibiting promising features for H2 energy industries. Overall, the synthetic method of this study proposes advanced strategies for development of sensing materials with maximized gas sensing performance.