Real-Time Smoke Detection in Surveillance Videos Using an Enhanced RT-DETR Framework with Triplet Attention and HS-FPN by Lanyan Yang, Yuanhang Cheng, Fang Xu, Boning Li, Xiaoxu Li

“…Our system integrates advanced modules, including triplet attention, ADown, and a high-level screening-feature fusion pyramid network (HS-FPN), to address challenges related to low-resolution imagery, real-time processing constraints, and environmental variability. …”
Get full text

Research on a Quality Control Method for L Band Second-Level Radiosonde toward Assimilation Applications by Dan WANG, Jincheng WANG, Weihong TIAN

“…Since 2003， the World Meteorological Organization （WMO） proposed and coordinated a switch from traditional radiosonde format （TAC code） to second-resolution binary （BUFR） report， which leads the assimilation of second radiosonde data in numerical prediction model becoming the future trend.Data quality control is a key and basic work before assimilation application.In addition， the observational error characteristics of Lband second-level radiosonde are significantly different from that of conventional radiosonde.Therefore， a two-step quality control method for assimilation application is developed in this paper.The first step integrates conventional radiosonde quality control procedures and inserts several additional steps according to the data characteristics of second-level radiosonde， which aims at eliminating human-observation errors as much as possible.The second step introduces a dynamic blacklist checking module into the assimilation system， at the same time remaining the old OMB （observation minus background） check method， so as to control the representative deviation between the observation and the model background.By comparing and analyzing the statistical characteristics of the observation samples before and after the quality control procedure， and combining them with the actual assimilation analysis results of the NWP model， the rationality of the two-step quality control procedure has been comprehensively verified.It reveals that the data anomaly is greatly reduced after the two-step quality control procedure， at the same time the OMB is closer to a Gaussian distribution.Moreover， the validity of data assimilation is enhanced leading to a better assimilation analysis effect.The work lays a foundation for the future operational application of L-band second-level radiosonde data in GRAPES （Global/Regional Assimilation and Prediction System） model.…”
Get full text

Bioinformatics analysis identifies ZBTB16 as a potential immune biomarker for lung cancer and pan-cancer by Danfei Shi, Yunxiang Cai, Di Zhu, Xinmin Li, Yong Li

“…We conducted a comprehensive bioinformatics analysis of these differentially expressed genes, aiming to explore their enrichment functions and pathways, relative expression levels, interaction networks, and weighted gene co expression network (WGCNA) module analysis. …”
Get full text

PFO‐ACCESS: Augmenting Communications for Medical Care or Closure in the Evaluation of Patients With Stroke With Cardiac Shunts by Yasaman Pirahanchi, Benjamin Shifflett, Ehtisham Mahmud, Stefanie Brennan, Dawn M. Meyer, Melissa Mortin, Lovella Hailey, Jeffrey Bowers, Vikas Ravi, Julián Carrión, Reza Bavarsad, Kunal Agrawal, Royya Modir, Thomas Hemmen, Brett C. Meyer

“…The aim of the PFO‐ACCESS (Augmenting Communications for Medical Care or Closure in the Evaluation of Stroke Patients With Cardiac Shunts) program (which included implementation of the Viz.ai PFO‐specific communications module) was to determine if there was any change in PFO management due to improved communication between stroke and interventional cardiology teams. …”
Get full text

Automated severity level estimation of wheat rust using an EfficientNet-CBAM hybrid model by Sapna Nigam, Rajni Jain, Vaibhav Kumar Singh, Ashish Kumar Singh, Hari Krishna, Hari Krishna

“…The convolutional Block Attention Module was integrated into EfficientNet-B0 in place of the SE module to enhance feature extraction by simultaneously considering channel and spatial information. …”
Get full text

Two-Stream Modality-Based Deep Learning Approach for Enhanced Two-Person Human Interaction Recognition in Videos by Hemel Sharker Akash, Md Abdur Rahim, Abu Saleh Musa Miah, Hyoun-Sup Lee, Si-Woong Jang, Jungpil Shin

“…In the study, we proposed a two-stream deep learning-based HIR system to address these challenges and improve the accuracy and reliability of HIR systems. …”
Get full text

Power Grid Load Forecasting Using a CNN-LSTM Network Based on a Multi-Modal Attention Mechanism by Wangyong Guo, Shijin Liu, Liguo Weng, Xingyu Liang

“…The Channel Attention module is then applied to weight different feature channels, highlighting important information and reducing redundancy. …”
Get full text

Complex-Scene SAR Aircraft Recognition Combining Attention Mechanism and Inner Convolution Operator by Wansi Liu, Huan Wang, Jiapeng Duan, Lixiang Cao, Teng Feng, Xiaomin Tian

“…By integrating the MTCN module and involution, performance is enhanced. The Multi-TASP-Conv network (MTCN) module aims to effectively extract low-level semantic and spatial information using a shared lightweight attention gate structure to achieve cross-dimensional interaction between “channels and space” with very few parameters, capturing the dependencies among multiple dimensions and improving feature representation ability. …”
Get full text

The Assessment of Spinal Alignment Based on a Computer-Assisted Electromechanical Device by Israa Khalifa, Aseel Ghazwan, Luay Asaad Mahmood

“…The spine plays a crucial role in maintaining the biomechanical functionality of the skeletal system. It protects the spinal cord and facilitates movement, among other vital functions. …”
Get full text

Index Method of Evaluating the Performance of Economic Activities by A. Sh. Kamaletdinov, A. A. Ksenofontov

“…Data analysis and parameter estimation were carried out by means of the statistical information processing program IBM SPSS Statistics 20, the analytical module of the Russian Taxes regional tax revenue information analysis system and the MS Excel 365 tabular processor. …”
Get full text

Designing an AI-Supported Framework for Literary Text Adaptation in Primary Classrooms by Savvas A. Chatzichristofis, Alexandros Tsopozidis, Avgousta Kyriakidou-Zacharoudiou, Salomi Evripidou, Angelos Amanatiadis

“…Guided by multiliteracies theory, Vygotskian pedagogy, and epistemic justice, the system aims to enhance interpretive literacy, developmental alignment, and cultural responsiveness among learners aged 7–12. …”
Get full text

Empowering Energy Transition: IoT-Driven Heat Pump Management for Optimal Thermal Comfort by Ivica Glavan, Ivan Gospić, Igor Poljak

“…This paper analyzes the process of energy transition from traditional solid fuel heating to an air-to-air (A2A) heat pump-based heating system. Special emphasis was placed on the implementation of new technologies for improved management of energy systems, aiming to elevate both comfort levels and energy efficiency. …”
Get full text

Semantic segmentation of substation tools using an improved ICNet network by Guozhong Liu, Qiongping Tang, Changnian Lin, An Xu, Chonglong Lin, Hao Meng, Mengyu Ruan, Wei Jin

Get full text

HARDWARE AND SOFTWARE COMPLEX FOR FUNCTIONAL STATE MONITORING OF MOTHER AND FETUS by I. V. Tolmachyov, K. S. Brazovsky, A. S. Tsverova, V. S. Ripenko

“…At the moment in the world there are no complexes which enable monitoring of the fetus without any risk to his health. So the main aim of this work is development of hardware and software system to assess the functional status of the mother and fetus by the analysis of the signals recorded from the abdominal electrodes.Objectives:1) to formulate the requirements to noninvasive monitoring system and functional state assessment of mother and fetus.2) to develop algorithms and software for monitoring system and functional state assessment of mother and fetus.3) to determine fetus functional state at 32–33 week in women with uncomplicated pregnancies and in women with placental insufficiency in the compensation stage and subcompensation.The system concept is based on fetal heart rate recognition and analysis. …”
Get full text

Study on the Influence of Mechanical Spindle Axial Movement based on ADAMS Simulation Analysis by Jia Yang, Wu Yalong, Wang Zhongmin, Mao Zhiwei, Zhang Jinjie

“…By using Creo,ADAMS software,the gear transmission system dynamics simulation model is built and the no axial movement,one-way axial movement and reciprocating axial movement scenarios is simulated. …”
Get full text

M-APSK phase detection algorithm and parallel carrier synchronization method by Hao HUAN, Kexue REN

“…In order to realize the precise phase correction of high order modulation in M-APSK modem, the Q-power non-data aided phase detection method recommended by DVB-S2 standard for 16APSK and 32APSK was extended to 64APSK, 128APSK and 256APSK.An improved algorithm was proposed to solve the problem of unstable loop operation when the proportion of constellation points used for phase detection in high-order modulation was low.By using threshold judgment on the amplitude of the received symbol after power normalization, phase detection was performed only when the amplitude was higher than the threshold, and the filter state and phase compensation value were not changed when the amplitude was lower than the threshold, so as to improve the phase detection effectiveness and reliability of constellation points and reduce the lock threshold.Aiming at the problem that the symbol rate of high-speed data transmission was very high but the working clock frequency of the processor was relatively low, a parallel carrier synchronization method for M-APSK was proposed, which could meet the needs of the receiver’s working clock.Compared with the traditional constant coding and modulation (CCM) carrier synchronization loop, the parallel method could also be applied to the frequency tracking of variable coding and modulation (VCM) system.…”
Get full text

Rail Transit Prediction Based on Multi-View Graph Attention Networks by Li Wang, Xin Wang, Jiao Wang

“…Traffic prediction is the cornerstone of intelligent transportation system. In recent years, graph neural network has become the mainstream traffic prediction method due to its excellent processing ability of unstructured data. …”
Get full text

SIGNSENSEI: ADVANCING UNIVERSAL COMMUNICATION THROUGH DEEP LEARNING by Prabhjot Kaur, Anupama Kaushik, Apoorv Mohit, Aryan Anand, Jasneet Kaur Bhasin

“…This paper explores the comprehensive development, implementation, and impact of SignSensei. SignSensei aims to empower users by allowing them to train the system with their own words on the website. …”
Get full text

Blockchain based intelligent disbursement in National Scholarship Portal by Lifna Challissery Samu, Neelkanth Khithani, Kushl Alve, Vedang Gambhire, Atharva Hande, Shivam Choubey

“…Thus, the implementation of the proposed system aims to alleviate the financial insecurities faced by students during their studies, simplify their search for scholarship opportunities, and enable them to focus more on their academic pursuits.…”
Get full text

Dual-robot formation transport in random environment and narrow restricted area via improved DDPG navigation by Liang Tang, Ronggeng Ma, Bowen Chen, Yisen Niu

“…Abstract On the background of two robots transporting long and no deflection deformation rods, a dual-robot system model is established, and via optimized navigation method based on Deep Deterministic Policy Gradient (DDPG) to address the issues of poor navigation paths, and resolved the robot system encountered the blocked exits in narrow areas during collaborative transportation. …”
Get full text